JP2005049212A - Print quality inspection device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform proper inspection even displacement or enlargement/reduction occurs in each image at the printing time, in inspection of a print formed by printing a discrete image on a form where a fixed-form image is printed. <P>SOLUTION: An inspection image acquired by reading the print is projected in the horizontal and vertical directions, to thereby determine the projection waveform in each direction (S12). A geometric transformation parameter between the projection waveforms of both images is calculated by using Hough transform so that a characteristic peak of the projection waveform is adapted to a characteristic peak of the projection image in the same direction of the fixed-form image (S14). The fixed-form image is aligned to the inspection image by the parameter, and a differential waveform between the projection waveforms of both images is calculated (S20). A transformation parameter between both images is calculated so that a peak of the differential waveform is adapted to a peak of the projection waveform in the same direction of the discrete image (S22). The fixed-form image and the discrete image are transformed by each corresponding transformation parameter and then synthesized (S28), and a synthesized image acquired resultantly is collated with the inspection image (S30). <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a print quality inspection apparatus for inspecting the print quality of a printed material, and more particularly to an apparatus for inspecting a printed material formed by a plurality of printing processes.

  When printing a standard document such as an invoice or delivery note, prepare a sheet of paper with a standard part printed in advance (referred to as pre-printed paper). Amount of money) is often printed individually on the preprinted paper.

  As apparatuses for inspecting the quality of printing results for such preprinted paper, there are apparatuses disclosed in Patent Document 1 and Patent Document 2.

  The apparatus disclosed in Patent Document 1 stores an image of a standard part of preprinted paper, takes a logical OR for each pixel of the image of the standard part and individual information printed on the preprinted paper, and prints the result. Is compared with the logical sum image to determine whether the print result is good or bad.

  However, in the processing method of this apparatus, when positional deviation (parallel movement) or magnification fluctuation (enlargement / reduction) occurs when an image of individual information is printed on preprinted paper, the accuracy of pattern matching decreases, Pattern matching itself may be impossible. For example, if the positional deviation or magnification fluctuation is within the allowable range, the print quality should be determined to be good, but may be determined to be poor due to a pattern matching defect. Further, there may be a positional shift in the printing of the image of the standard part of the preprinted paper. In this case, the same problem occurs.

  Further, the apparatus disclosed in Patent Document 2 masks a portion corresponding to the image of the standard portion of the preprinted paper from the image obtained by reading the print result, and compares the image remaining as a result of the mask with the image of the individual information. Then, the quality of the print result is determined.

  In the processing method of this apparatus, if there is a positional deviation or the like in the printing of the standard part of the preprinted paper, the printed standard part cannot be completely masked, which has an adverse effect on the quality determination. In addition, if white paper is mixed in the preprinted paper bundle of the paper supply unit, the same mask result is obtained for both white paper and preprinted paper, so the quality is good even when individual information is printed on white paper. It will be judged.

JP-A-3-281276 JP-A-11-78183

  In the quality inspection of a printed matter formed by a plurality of printing processes, such as when printing on preprinted paper, the present invention can be applied to geometric images such as misalignment, enlargement / reduction, etc. The purpose is to enable proper quality judgment even when there is a change.

  The print quality inspection apparatus according to the present invention is a print quality inspection apparatus for inspecting a printed material in which a first reference image and a second reference image are printed by separate printing processes on the same sheet surface. Calculating a first geometric transformation between the first reference image and a first corresponding image portion corresponding to the first reference image in the printed surface image obtained by reading the printed surface of the printed matter; A second conversion unit that calculates a second geometric transformation between the first conversion calculation unit, the second reference image, and a second corresponding image portion corresponding to the second reference image in the print surface image; A first determination for determining the quality of the print surface image using a conversion calculation unit, the first reference image, the second reference image, the first geometric conversion, and the second geometric conversion. A section.

  In one aspect of the present invention, the first conversion calculation unit calculates the first geometric conversion by comparing predetermined image features of the print surface image and the first reference image.

  According to another aspect of the present invention, the second conversion calculation unit reflects the first geometric transformation on the predetermined image feature of the first reference image and the predetermined image of the print surface image. The second geometric transformation is calculated by comparing the difference with the image feature and the predetermined image feature of the second reference image.

  According to still another aspect of the present invention, the second conversion calculation unit reflects the inverse of the first geometric conversion in the predetermined image feature of the print surface image and the first reference. The second geometric transformation is calculated by comparing the difference of the image with the predetermined image feature and the predetermined image feature of the second reference image.

  Here, as the predetermined image feature, for example, a projection waveform obtained by projecting an image in a predetermined direction within the plane of the image can be used. In consideration of misalignment and enlargement / reduction of a two-dimensional image, it is desirable to use projection waveforms in two or more different directions.

  In one preferable aspect of the present invention, when the first conversion calculation unit and / or the second conversion calculation unit compare the projection waveforms of two images, a plurality of feature points of the respective projection waveforms are best. The associated geometric transformation is calculated using the Hough transform operation.

  The apparatus according to the present invention can be suitably applied, for example, when the first reference image is an image including a ruled line, a table ruled line, or a frame line, but the application target is not limited to this. If the first reference image is an image having a pattern in which black pixels are specifically densely arranged along a specific direction, even if it is not a ruled line, a table ruled line or a frame line, By performing projection to create a projection waveform, geometric transformation can be calculated in the same manner as described above.

  In another preferred aspect of the present invention, the first determination unit includes a first post-conversion image obtained by applying the first geometric transformation to the first reference image, and the second geometry. A second converted image obtained by performing a geometrical conversion on the second reference image to form a combined image, and based on the comparison between the printed surface image and the combined image, the printed surface image Determine the quality of the.

  In still another preferred aspect, when the geometric transformation calculated by the first transformation calculation unit or the second transformation calculation unit deviates from a predetermined allowable range, the print surface image is defective. The determination is made so that the determination process by the first determination unit is not executed.

  Hereinafter, the best mode for carrying out the present invention (hereinafter referred to as an embodiment) will be described with reference to the drawings.

  First, referring to FIGS. 1A to 5, a verification process executed by the print quality inspection apparatus according to the present embodiment will be described. The image matching process of the present embodiment is a process for inspecting the quality of a print result of a standard document composed of a standard information part and an individual information part.

  In the printing of a standard document, it is often the case that a preprint paper on which a standard information portion is printed in advance is prepared and the individual information portion is printed on the preprint paper. Since the fixed information portion has the same contents, the printing cost per sheet can be reduced by creating a printing plate and printing in large quantities. On the other hand, since the individual information portion differs for each document, it is necessary to print it individually.

  In the example shown in FIGS. 1A to 5, “receipt” is taken as an example of the standard document. In the case of the illustrated “receipt”, as shown in FIG. 1A, the standard image 120 which is an image of the standard information part is a title 122 of the document, a title 124 attached to the destination, a monetary entry column 126, and issuance of the receipt An image element indicating the original company name 128 is included. The character string “receipt” of the title 122 is underlined for emphasis. Further, a thick frame line 126a is shown in the amount entry column 126 for emphasis. In addition, the individual image 140 that is an image of the individual information portion includes image elements indicating the destination name 142 and the receipt amount 144 as shown in FIG. 1B. The inspection image 100 indicates an image obtained by reading a printed surface of a printed material on which the individual image 140 is printed on a preprinted paper on which the standard image 120 is printed in advance.

  When a single standard document is created by the two-stage printing process in which the individual image 140 is printed after the standard image 120 is printed, geometric distortion such as image misalignment, enlargement / reduction, etc. May occur. The inspection image 100 shown in FIGS. 1A and 1B is printed with the fixed image 120 magnifying in the enlargement direction and shifted to the left of the correct position, and the individual image 140 magnifying in the reduction direction and magnifying to the right of the correct position. The image of the printing surface when it is done is shown.

  In order to inspect the print result of such a standard document, the projection waveform of the standard image 120 is first obtained in the processing of this embodiment. When correcting geometric distortion of a two-dimensional image, it is necessary to obtain projection waveforms in at least two different directions and perform processing on the projection waveforms along these directions. However, since the contents of the processing performed on the projection waveform along each direction are the same, the processing on the projection waveform along one direction will be described below.

  FIG. 2A illustrates a projection waveform 240 along the horizontal (lateral) direction of the standard image 120, that is, the X direction.

  The projection of the image here is a process of integrating the values of the respective pixels on the path along the integration path set in the image. A series in which integration values (that is, projection values) of a plurality of integration paths set in an image are arranged according to the order of the integration paths is a projection waveform. The projection waveform along the X direction is a projection waveform obtained when the integration path is set for each pixel interval in parallel along the Y direction (that is, the vertical direction) perpendicular to the X direction.

  Similarly, a projection waveform is also obtained for the inspection image 100. FIG. 2B shows a projection waveform 220 along the X direction of the inspection image 100.

  Next, by comparing the projected waveform 210 of the fixed image 120 and the projected waveform 220 of the inspection image 100, a conversion parameter indicating a geometric conversion that occurs when the fixed image 12 is printed is calculated. In this calculation, as shown in FIG. 3, characteristic peaks 212 (shown by black circles) and 222 (shown by white circles) are extracted from the projected waveform 210 of the fixed image 120 and the projected waveform 220 of the inspection waveform 100, respectively, and the Hough Using the concept of (Hough) conversion, conversion parameters are calculated such that the X-direction positions of the corresponding peaks 212 and 222 coincide with each other.

  As a rule for selecting characteristic peaks 212 and 222 from the projection waveforms 210 and 220, for example, a peak in which a difference in wave height (integral value) from a position immediately before or immediately after in the projection waveform is larger than a predetermined threshold is selected. And the rules. However, this is only an example, and other rules can be used.

  According to the calculation process using the Hough transform, the amount of positional deviation in the X direction between the projection waveforms 210 and 220 and the enlargement or reduction magnification in the X direction can be obtained as conversion parameters. The conversion parameter of the geometric conversion obtained by this processing indicates the amount of positional deviation in the X direction and the enlargement / reduction magnification that occur when the standard image 120 is printed. In this example, for example, it is required that the amount of positional deviation in the X direction when printing the standard image 120 is 5 pixels in the left direction, and the enlargement / reduction ratio is 1.05 (enlargement). Details of the calculation process using the Hough transform will be described later.

  After the conversion parameters are calculated in this way, next, as shown in FIG. 4A, the projection waveform 210 of the standard image 120 is translated and enlarged / reduced (scale conversion) by using the conversion parameters. The position is aligned with the projected waveform 220 of the inspection image 100. Then, the differential waveform 230 is obtained by subtracting the aligned projection waveform 210 from the projection waveform 220. This difference waveform 230 is equal to the projection waveform along the X direction of the difference image obtained by subtracting the fixed image 120 corrected for positional deviation and enlargement / reduction in the X direction from the inspection image 100.

  That is, for example, if the above processing is performed in each of the X and Y directions, the positional deviation and enlargement / reduction in both the X and Y directions when the standard image 120 is printed can be corrected. Therefore, when the corrected standard image 120 is subtracted from the inspection image 100, only the image portion corresponding to the individual image 140 remains in the inspection image 100. The differential waveform 230 is equivalent to the projection waveform along the X direction of the image portion corresponding to this individual image.

  Next, as shown in FIG. 4B, a projection waveform 240 along the X direction of the individual image 140 is obtained.

  Next, by comparing the projected waveform 240 and the differential waveform 230 of the individual image 140, the geometric transformation parameters generated when the image 12 is printed are calculated. In this calculation, the Hough transform is used in the same manner as the calculation of the conversion parameter between the inspection image 100 and the fixed image 120, and the characteristic peaks 242 (indicated by black circles in FIG. 4B) and 232 are obtained from the respective waveforms 240 and 230. (Indicated by white circles in FIG. 4A) is extracted, and conversion parameters are calculated such that the X-direction positions of the corresponding peaks 242 and 232 coincide with each other by using the idea of the Hough transform. In this example, for example, it is required that the amount of positional deviation in the X direction when printing the individual image 140 is 5 pixels in the right direction and the enlargement / reduction magnification is 0.95.

  The rules for selecting characteristic peaks 242 and 232 from the peaks of the waveforms 240 and 230 can be the same as those for the projection waveforms 210 and 220. However, the threshold value for the difference in wave height, which serves as a reference for selection, can be different from that for the projection waveforms 210 and 220.

  When the geometric conversion parameters at the time of printing of the individual image 140 are calculated in this way, next, as shown in FIG. 5, the fixed image 120 and the individual image 140 are divided by the respective conversion parameters. After the correction, the fixed image 120 and the individual image 140 after the correction are added for each pixel to create a composite image 160. Then, the composite image 160 and the inspection image 100 are collated to determine whether the printed matter has a defect. The composite image 160 shows an ideal print surface state when ignoring the positional deviation and enlargement / reduction at the time of printing. By comparing this with the inspection image 10, the defect of the actual print surface is obtained. be able to. For example, when a certain pixel is black on the composite image 160 but white on the inspection image 100, the pixel may be a missing pixel. Conversely, a pixel that is white on the composite image 160 but black on the inspection image 100 may be a stain during printing.

  Even if there is a difference between the composite image 160 and the inspection image 100 as described above, the difference is not always an image defect immediately. This is because there is no problem in terms of quality unless a cluster of pixels with a difference is noticeable. Therefore, in the collation process, a block of pixels indicating missing print or smudge is identified from the comparison of both images, and it is determined whether or not each block is a defect based on the size and darkness of the block. Various existing methods such as pattern matching can be used for the collation process between the composite image 160 and the inspection image 100.

  The flow of processing executed by the apparatus according to this embodiment has been described above using specific examples. In the above description, a simple example in which the geometric distortion at the time of printing the standard image 120 and the individual image 140 is only the positional deviation in the X direction is used. Therefore, only the processing in the X direction as described above is used. Was enough. However, in general, there is a possibility that displacement or enlargement / reduction may occur in the Y direction during printing. Therefore, in practice, the same calculation process as described above is performed for the Y direction to obtain the positional deviation and enlargement / reduction of the fixed image 120 and the individual image 140 in the Y direction, and when the composite image 160 is created. , X, Y direction displacement and enlargement / reduction are considered.

  The order of processing shown in the above description is merely an example, and other orders are possible. For example, the fixed image 120 can be acquired at the stage where printing of a series of fixed document groups is started, so that the projection waveform 210 can be created at that time. Therefore, the projection waveform 210 can be created and stored at that time, and the projection waveform 210 can be read and used when inspecting each inspection image 100. Further, since the individual image 140 can also be acquired at the time when printing of each standard document is started, the projection waveform 240 can be created at that time.

  As described above, the print quality inspection apparatus according to the present embodiment creates a composite image 160 that corrects the positional deviation and enlargement / reduction of the standard image 120 and the individual image 140 during printing, and uses the composite image 160 as a template. As shown in FIG. In the case of each technique of Patent Documents 1 and 2, when the fixed image 120 and the individual image 140 are individually misaligned or enlarged or reduced, there is a possibility that collation accuracy may be reduced or collation failure may be caused. According to the processing method of this embodiment, such a problem is solved. Even when white paper is mixed in the preprint paper, it can be detected by inspection of the printing surface.

  In the actual print quality inspection, if the positional deviation or enlargement / reduction rate of the standard image 120 or the individual image 140 at the time of printing is slight within an allowable range, it is judged that the quality of the print result is good. This is beneficial in terms of productivity and printing costs. In this embodiment, it can be said that it is more useful than the prior art shown in patent documents 1 and 2 from such a viewpoint. If it is found that the positional deviation or enlargement / reduction of the standard image 120 or the individual image 140 during printing exceeds the allowable range, it may be determined that the printing result is defective.

  In the processing of the present embodiment described above, the fixed image 120 and the individual image 140 have such a relationship that the peak corresponding to the peak of the projected waveform 210 of the fixed image 120 appears significantly in the projected waveform 220 of the inspection image 100. If it does, it will be established.

  For example, in the case of business documents such as receipts, invoices, and invoices, the fixed image 120 often includes ruled lines, table ruled lines, frame lines for emphasis, and the like. Since the ruled lines, the table ruled lines, and the frame lines are represented by densely arranging black pixels in the vertical and horizontal directions of the image, that is, in the X and Y directions, each of the regular images 120 along the Y direction and the X direction. In the projected waveform, significant peaks appear at positions corresponding to these lines. On the other hand, in the case of business documents, the individual image 140 is often composed of characters and simple patterns, and these characters and patterns are often arranged between the ruled lines, table ruled lines, and frame lines. In general, even if an image composed of characters and simple patterns is projected, the projection value does not become as large as a line in which black pixels are closely arranged. Therefore, significant peaks corresponding to lines such as ruled lines, table ruled lines, and frame lines also appear in the projected waveforms obtained by projecting the inspection image 100 in the X and Y directions. Therefore, the processing of this embodiment described above can be effectively applied to a document including ruled lines, table ruled lines, and frame lines.

  However, business documents including ruled lines, table ruled lines, frame lines, and the like are merely examples of objects to which the method of this embodiment can be applied. Even if the document does not include such ruled lines, any document can be used as long as the peak corresponding to the peak of the projected waveform of the fixed image 120 appears in the projected waveform of the inspection image 100. The above method can be applied.

  In addition, the direction along which the projection waveform is used in executing the processing of the present embodiment can be determined according to the property of the image to be inspected. For example, when inspecting business documents in which lines and broken lines in the X and Y directions are conspicuous, projection waveforms along the X and Y directions may be created as exemplified above. Similarly, if it is known in advance that the fixed image 120 has a pattern in which black pixels are specifically densely arranged along a specific direction, projection is performed in that direction, and the direction perpendicular to that direction is projected. By creating the projected waveform along the line, it is possible to calculate the amount of displacement and enlargement / reduction in the same manner as described above.

  Next, a specific example of processing using the Hough transform for obtaining the positional deviation between projection waveforms and the amount of enlargement / reduction will be described. This process is the same as the process described in Japanese Patent Application Laid-Open Nos. 2003-91730 and 2003-109007 by the applicant. In the following, a case where a conversion parameter between the projection waveform 220 of the inspection image 100 and the projection waveform 210 of the standard image 120 is obtained will be described as an example. However, conversion between the difference waveform 230 and the projection waveform 240 of the individual image 140 is described. The same process may be used when obtaining parameters.

  In this process, in order to obtain a transformation parameter for geometric transformation, a characteristic portion of each projection waveform is extracted, and the positional relationship is compared to obtain a transformation parameter. Here, a characteristic peak of the projected waveform is used as the characteristic part, but this is merely an example, and other characteristic points may be a differential waveform of the projected waveform such as an inflection point. Further, the position of the maximum value of the projection waveform, the position where the projection wave height becomes a predetermined value, the position of the center of gravity of the projection waveform, the width of the peak, etc. may be used in combination with the characteristic position and amount obtained from the projection waveform itself. . For the image data in the database, it is also preferable to extract these characteristic positions and amounts in advance and associate them with the corresponding image data.

  For example, when a characteristic peak of each projection waveform is obtained, the position of this characteristic peak is obtained. Since there are usually multiple characteristic peaks in the projected waveform, this position is a list of positions.

  Then, a conversion parameter is obtained by comparing the peak list related to the projection waveform 220 of the inspection image 100 and the peak list related to the projection waveform 210 of the standard image 120. In this comparison process, conversion parameters are obtained by associating each peak position. For example, there are three peaks (A, B, C and α, β, γ) in two compared projection waveforms. (Refer to FIG. 6A and FIG. 6B) When two image data are the same image data, when A and α and B and β are matched (correct), A and β and B The situation is different from the case of matching (incorrect) matching γ as follows. In the following description, for the sake of simplicity, it is assumed that the scales of the two image data to be compared are the same (that is, scaling (enlargement / reduction) is not considered). In FIG. 6A and FIG. 6B, the portions other than the peaks are shown flat for explanation.

  That is, the translation amount obtained from the relationship between A and α and the translation amount obtained from the relationship between B and β substantially coincide with each other, whereas the translation amount obtained from the relationship between A and β The amount of translation required by the relationship between B and γ is generally different.

  Therefore, an appropriate value is obtained by trying all associations brute force and comparing each result to obtain the most likely value (maximum likelihood value) by the majority rule.

  When scaling (enlargement / reduction) correction is required, the relationship between the scaling amount k and the parallel movement amount s is obtained from the correspondence between the peak positions in each peak list. That is, when associating the peak position xi related to one image data with the peak position ξi related to the other image data, the relationship of the following equation (1) is obtained.

  When a plurality of straight lines according to the formula (1) are drawn on the ks plane ((scaling correction amount)-(parallel movement amount) plane) in association with each peak position, ideally, a large number of straight lines. Is obtained, and the correct scaling correction amount is obtained as the coordinate value (k, s). Here, “ideally” is because k and s obtained from each straight line do not necessarily match due to factors such as calculation errors. Therefore, k and s are determined by selecting the maximum likelihood value based on the majority rule. Focusing on the fact that a large number of parameter candidates gather in the vicinity of a certain value when correctly associated in this way, the maximum likelihood value is obtained using the majority rule.

  Similarly, for example, the peak list of the projection waveform with respect to the horizontal direction of one image data is correctly associated with the peak list of the projection waveform with respect to the vertical direction of the other image data (that is, the parameter value having a larger number of parameter candidates). If it gathers in the vicinity), it can be determined that it has been rotated 90 degrees. Further, if the order of one peak list is changed (in other words, the direction of one projection waveform is reversed), it can be determined that the peak list is reversed. That is, this process can be applied to the case where the standard image 120 and the inspection image 100 are rotated by 90 degrees or turned over. That is, primary conversion parameters including rotation, inversion, scaling, and the like can be obtained by similar processing. In the following description, only the scaling correction and the parallel movement will be described in order to simplify the description.

  In the above description, the projection waveform 220 of the inspection image 100 and the projection waveform 210 of the standard image 120 are compared. However, the differential waveforms obtained by differentiating the respective waveforms may be compared. When a copied document is used as the inspection image 100, if the image data includes a halftone area such as a photograph or drawing, the density value often changes greatly, and the projection peak value may not be stable. Many. Even in such a case, the projections and depressions of the projection waveform are maintained in many cases, so it is often preferable to use the differential projection waveform as a comparison target.

  Furthermore, here, the projection waveform itself is used to create the peak list from the projection waveform. However, the projection waveform generated from the character string has many high-frequency components, and also contains many noise components due to writing and imprinting. On the other hand, when estimating the transformation parameters, it is sufficient to use information such as the pitch of the rows and the length of each row, and the shape such as the pitch of these rows is reflected in the low-frequency component of the projection waveform. Therefore, low-pass filtering of the projected waveform is performed from the viewpoint of noise removal and stabilization of characteristic peak detection processing. In addition, because the projection waveform that has been low-pass filtered still has minute irregularities, a large number of characteristic peaks are detected and the amount of computation increases, so the global irregularities necessary to estimate the conversion parameters are reduced. It is preferable to include only the position of the characteristic peak to represent in the peak list.

  Therefore, an outline of a method for detecting only characteristic peaks representing global unevenness will be described below. FIG. 7 is a diagram for explaining a characteristic peak detection method for detecting the general unevenness of the projection waveform. In FIG. 7, p (x) is a peak value at the position x of the projection waveform after low-pass filtering, and α is a predetermined window width. At the characteristic peak position, the slope of the projection waveform becomes zero, but instead of obtaining this as the zero crossing point of the differential projection waveform, the slope is obtained by the slope of the straight line connecting the projection wave height values at both end positions of the window. The slope of this straight line at the position Xa is proportional to S (Xa) in the following equation (2).

  The point where the sign of S (Xa) changes is the characteristic peak position. Here, it is preferable that information indicating whether the point is a convex peak changing from positive to negative or a concave peak changing from negative to positive is acquired together with the peak list as peak attribute information. By referring to the projected wave height at a point a certain distance α from the target position in this way, the peak representing the general unevenness of the projected waveform without being affected by the minute unevenness of the projected waveform Candidates can be detected.

  From these peak candidates, only effective peaks for the detection of conversion parameters are selected. Since the conversion parameter is obtained by associating the general unevenness of the projection waveform, it is better to remove the minute peak.

  FIG. 8 is a diagram showing the operating principle of the means for selecting characteristic peaks. In FIG. 8, the positions of characteristic peak candidates (a total of five points from x1 to x5) obtained by the method shown in FIG. 7 are detected. In the example of FIG. 8, for example, characteristic peaks at x2 and x3 are peak candidates to be excluded as minute irregularities.

  In this embodiment, the following processing is performed to exclude a minute characteristic peak. That is, for a certain characteristic peak position Xn, a quantity Dn related to the difference between the projected wave height value at that position and the wave height value at the adjacent characteristic peak position is obtained as in the following equation (3).

  In addition, for those having characteristic peaks adjacent to only one side, such as positions x1 and x5, the absolute value with respect to the difference from the peak value of the characteristic peak is Dn.

  When Dn thus obtained and the predetermined threshold value Td satisfy the condition of the following equation (4), the characteristic peak at the position Xn is actually added to the peak list as a peak for comparison.

  Further, the peak selection may be performed using the ratio Rn of the peak values as shown in the equation (5) instead of the difference between the projected peak values at the adjacent characteristic peak positions.

  Then, when Rn obtained by the equation (5) and the predetermined threshold value Tr satisfy the condition of the equation (6), a characteristic peak at the position Xn is added to the peak list.

  Further, for the purpose of assisting comparison between peak lists, it is also preferable to obtain the width and center of gravity of the projected waveform as peak attribute information. These pieces of information are not essential for conversion parameter detection, but can be used for speeding up detection of conversion parameters and improving reliability as described later.

  When obtaining the peak width here, if the projection waveform does not contain noise and does not contain offset due to paper ground color, etc., the projection waveform is projected with the width of the non-zero section of the projection waveform. The width of the waveform may be used. However, since such an ideal example is extremely rare, in practice, the following processing must be performed in order to obtain the projection width.

  FIG. 9 is a diagram illustrating a process for detecting the width of the projection waveform. In the actual projection waveform, as shown in the areas A and B of FIG. 9, offsets due to noise, paper ground color, etc. (non-zero regions generated because the pixels are not partially “white” due to the ground color) )It is included. Therefore, threshold processing is performed to remove these components. However, when there are many crest values near the threshold value, a subtle difference in whether or not the threshold value is exceeded greatly affects the width of the detected projection waveform. Therefore, in the example of FIG. On the other hand, two different threshold values, Th and Tl, are set. Of these, the lower threshold value Tl is set so that most of noise and offset are less than this value, and the higher threshold value Th is set so that most of the original projection waveform exceeds this value. Is done.

  After setting such a threshold value, the width W of the projection waveform is obtained by the following equation (7).

  Here, as shown in FIG. 10, w (•) is “0” for x such that 0 <x <T1, “1” for x such that Th <x, and Tl <x <Th. Then, it is a function that linearly changes from “0” to “1”. Further, P represents the entire projection waveform, and therefore the expression (7) means that the summation is performed over the entire projection waveform. By providing a threshold value processing function having a gentle slope between the threshold values Th and Tl in this way, even when there are many projection waveforms near the threshold value that are problematic when a single threshold value is used. Thus, it is possible to prevent a significant difference from occurring in the width of the detected projection waveform.

  On the other hand, the center-of-gravity position G of the projection waveform is obtained from the following equation (8).

  A conversion parameter is obtained using the peak list and peak attribute information thus obtained. Next, the procedure for obtaining the conversion parameter will be outlined. First, referring to the respective peak lists of the projection waveform 220 of the inspection image 100 and the projection waveform 210 of the standard image 120, the peak positions are associated. The peak positions are associated with each other for the convex peak and the concave peak with reference to the peak attribute information. Next, based on the fact that the scaling correction amount and the parallel movement amount required for matching the two characteristic peak positions associated with each other have a linear relationship, the (scaling correction amount)-(parallel movement amount) plane Perform voting on a straight line. When such a voting process is performed for all combinations of characteristic peak positions, a distribution of accumulated voting values is formed on the (scaling correction amount)-(parallel movement amount) plane. On the (Scaling correction amount)-(Parallel movement amount) plane, the point at which this cumulative vote value increases indicates that the two characteristic peak lists are in good agreement. , Parallel movement amount) is a conversion parameter to be obtained.

  That is, as shown in FIG. 11, the characteristic convex peak list (A1 to A3) for the projection waveform 210 of the standard image 120 and the characteristic convex peak list (B1 to B4) obtained from the projection waveform 220 of the inspection image 100 are displayed. If there is, the peak positions are associated with each other for the two characteristic peak lists. For example, when associating the characteristic peaks A1 and B1, if the scaling correction amount necessary for matching the peak positions of the two is k and the translation amount is s, the coordinate “10” of A1 and the coordinate of B1 Using “19”, the following equation (9) is established.

  Since this equation (9) is a linear equation relating to k and s, a straight line can be drawn on the ks plane, that is, the (scaling correction amount)-(parallel movement amount) plane. If k and s constrained by the condition of equation (9) are used, both image data relating to the comparison can be superimposed at least at the characteristic peak positions A1 and B1. Therefore, “put a vote” on this straight line as a candidate for the correction amount. That is, voting is performed for each point on the straight line. I want to find the point with the most votes later as the maximum likelihood point.

  Similarly, by associating the characteristic peak A1 shown in FIG. 11 with other B2 to B4, in the (scaling correction amount)-(parallel movement amount) plane, with respect to A1, there are four straight lines. Will vote. In general, the position of the i-th peak in the characteristic peak list obtained from the projected waveform of image data in a database is Aix, and the position of the j-th peak in the characteristic peak list for the projected waveform of key image data is Bjx. Then, the relationship between the scaling correction amount k and the parallel movement amount s necessary for the peak positions of the two to coincide is the relationship of Equation (10), as in Equation (1).

  By performing the same voting process for the remaining characteristic peaks A2 and A3, a distribution of accumulated voting values is formed on the (scaling correction amount)-(parallel movement amount) plane.

  In addition to always casting one vote as in the above example, the vote value reflecting the peak value information of the characteristic peak can also be used. Regarding characteristic peaks, it is natural to associate large peak values or small peak values, so that the vote value is also changed according to the peak value. That is, the peak value of the i-th peak in the characteristic peak list obtained from the projection waveform is Ais, and the peak value of the j-th peak in the characteristic peak list for the projection waveform of the key image data is Bjs, for example, a vote value V is defined as the following equation (11).

  In order to apply the vote value of the equation (11), it is necessary to perform a normalization process in advance such that the maximum value of the peak value of the characteristic peak becomes a predetermined value.

  Thus, the cumulative vote value formed on the (scaling correction amount)-(parallel movement amount) plane reflects the degree of coincidence between the two characteristic peak lists. Therefore, if the peak point of the cumulative vote value formed on the same plane is searched, the coordinates of that point are obtained (scaling correction amount, parallel movement amount). By applying correction to one projection waveform using this correction amount, the two projection waveforms are best matched. In the example of FIG. 11, since (scaling correction amount, parallel movement amount) = (0.9, 10) has a maximum cumulative vote value of (cumulative vote value) = 3, the projection waveform of the fixed image 120 is 0.9. By multiplying and further translating by 10, the two projected waveforms are best matched.

  By the way, when the above correction amount is applied, it can be seen that only the characteristic peak B3 on the inspection image 100 side does not have a characteristic peak corresponding to the fixed image 120 side. From this, it is presumed that the characteristic peak B3 is a peak component due to noise such as writing on a document and imprinting. In the conversion parameter detection procedure of the present embodiment, even if such a characteristic peak due to noise or a detection omission occurs, the conversion parameter is estimated based on the majority rule called voting processing. A decrease in accuracy of the correction amount can be prevented.

  In the above description, the conversion parameter is detected by searching for the peak point of the cumulative vote value formed on the (scaling correction amount)-(parallel movement amount) plane. This is an “ideal” example. Actually, it is rare that a group of straight lines for voting exactly meet at one point due to an error at the time of detecting a characteristic peak or the printing accuracy of the original document. Therefore, as a procedure for estimating the transformation parameter, the (scaling correction amount)-(parallel movement amount) plane is divided into “quantization” into cells divided into n in the scaling correction amount direction and m in the parallel movement amount direction as shown in FIG. In other words, a method may be considered in which the total of the vote values of each cell is calculated and the cell having the maximum total is obtained. Then, the maximum likelihood value of the conversion parameter is determined by using the center coordinates (kc, sc) of the cell in which the cumulative total is maximum. In this way, it can be adapted to computer processing.

  In order to further improve the accuracy, as shown in FIG. 13, after obtaining a cell having the maximum total vote value, only the straight line group passing through the cell is selected from the straight line group to be voted, and those straight lines are selected. The coordinates of the point located at the shortest distance from the group are obtained, and the coordinates are set as the maximum likelihood value. The coordinates of such a point can be obtained by the following procedure.

  First, it is assumed that there are h straight line groups passing through a cell having the largest total vote value, and these are represented by the equation (12).

  Here, i = 0, 1,..., H−1. Between the distance di between one of these straight lines and a point (k, s) on the (scaling correction amount)-(parallel movement amount) plane, there is a relationship of the following equation (13).

  The square sum D of the distance from the point (k, s) to each straight line is expressed by the following equation (14).

  Therefore, the coordinates (kl, sl) of the point located at the shortest distance from the straight line group passing through the cell having the largest total vote value are obtained by solving the following equations (15) and (16) simultaneously. be able to.

  Furthermore, when obtaining the coordinates of the point located at the shortest distance from the straight line group, the distance from each straight line may be weighted by the vote value V defined in the equation (11). By performing weighting with the voting value V, correction amount detection is performed with emphasis on straight lines that are highly likely to be associated with characteristic peaks. In this case, equation (14) is changed to equation (17).

  In the equation (17), Vi is a vote value on the i-th straight line. Of the peak attribute information, the width and centroid information of the projected waveform can be used for speeding up the processing and improving the reliability in estimating the conversion parameter.

  That is, parameter estimation by voting generally has a so-called false peak problem that causes a relatively large peak at an inappropriate position. In order to reduce this effect, it is effective not to vote for the entire voting space, but not to vote for areas that are clearly deemed unnecessary. In the present embodiment, by limiting the voting area using the information on the width and the center of gravity of the projection waveform, the conversion parameter detection process can be speeded up, and the effect due to the above problem can be reduced.

  Specifically, the voting area in the scaling correction amount direction on the (scaling correction amount)-(parallel movement amount) plane can be limited by using the width of the projection waveform. When the projection waveform width of the image data in the database is Wr and the projection waveform width of the key image data is Wi, and the projection width ratio R shown in the following equation (18) is obtained, the scaling correction amount in the conversion parameter is close to R. Should be a value. This is because if two documents are the same document, their projection waveforms should be similar, and the difference in scaling should be the ratio of the projection width.

  Although the width of the projection waveform includes an error due to an offset due to noise, paper ground color, etc., the true correction amount, that is, the scaling coordinate of the peak position of the vote value is still expressed by equation (18). Should be close to R. Therefore, a predetermined scaling margin β centered on R in the equation (18) is set, and the voting area is limited within the range of the scaling margin β. In this case, among the straight lines drawn on the (scaling correction amount)-(parallel movement amount) plane, it is assumed that “voting” is performed within the range of the scaling margin β, and “voting” is performed outside the range. do not do.

  The voting area can also be limited from the center of gravity position information of the projection waveform. That is, if the centroid position obtained by the feature extraction means is accurate, the relationship of the following expression (19) should be established between the centroid positions of the two projection waveforms as in the expression (10). is there.

  In Equation (19), Gr and Gi are the projected centroid position of the image data in the database and the projected centroid position of the key image data, respectively. Actually, the obtained information on the center of gravity position includes an error due to an offset due to noise, paper ground color, etc. However, the true correction amount, that is, the peak position of the vote value is still ( It should be located in the vicinity of the straight line represented by the equation (19). Therefore, an appropriate parallel movement amount margin γ is set around the straight line of the equation (19), and the voting area is limited within the range.

  As a result, the voting area is limited to the range shown in FIG. 14, the voting process on a straight line outside this area or on a straight line that does not pass through this area is omitted, and the processing speed is increased. The effect of the problem can be reduced.

  It should be noted that, when interpolation processing is required, such as when a signal at a non-sampling position is required to perform projection waveform conversion processing based on the conversion parameters thus obtained, the nearest neighbor method, Various methods such as a linear interpolation method, a cubic convolution method, and a spline interpolation method can be used. Since the apparatus of the present embodiment does not particularly depend on a specific interpolation method, any of the interpolation methods listed above may be used. Since the above interpolation method is explained in many literatures, its explanation is omitted.

  As described above, the entire collation process by the print quality inspection apparatus of the present embodiment and the projection waveform matching process using the Hough transform used for the collation process have been described.

  Next, an example of a printing system to which the print quality inspection apparatus that executes the above-described processing is applied will be described with reference to FIG.

  In the printing system, an image is printed by the print processing unit 20 on the paper fed from the paper feed cassette 10, and the printed image is fixed on the paper by the fixing roller 25. In the case of an electrophotographic system, the print processing unit 20 includes a raster output scanner, a photosensitive drum, a developing device, a transfer roll, and the like. When printing a standard document, the preprinted paper on which the standard image 120 is printed is set in any of the paper feed cassettes 10, and the cassette 10 is selected as a paper supply source to start printing.

  A reading scanner 30 is provided downstream of the fixing roller 25 in the paper conveyance path. The printed surface of the printed material on which the image is fixed is optically read by the reading scanner 30. The reading scanner 30 irradiates the printing surface with an irradiation lamp 35, focuses the reflected light with a lens 36, and detects it with an imaging device 37. As the imaging device 37, for example, a line sensor in which cells are arranged in a line in a direction perpendicular to the paper conveyance direction in a plane parallel to the upper surface of the paper conveyance path can be used. By repeatedly reading the line sensor in synchronization with the sheet conveyance, a two-dimensional printed surface image can be acquired. When inspecting a color image, for example, a line sensor may be provided for each of R, G, and B colors. The image of the printing surface acquired by the imaging device 37, that is, the inspection image 100 is sent to the print quality inspection apparatus 50. The print quality inspection apparatus 50 executes the above-described verification process. The printed matter is discharged to the discharge tray 40 after the collation process.

  Here, according to the result of collation by the print quality inspection apparatus 50, the processing content for the printed matter after collation can be switched. For example, in addition to the non-defective product discharge tray, a defective product-dedicated tray is provided, and when the print quality inspection apparatus 50 determines that the printed product is defective, the printed product is discharged to the defective product-dedicated tray. It can also be done. In addition, a device for disposing of a defective printed material is provided on the paper conveyance path after collation, and when the printed quality inspection device 50 determines that the printed material is defective, the printed material can be disposed of by the disposal device. . As the disposal device, for example, a painting device for painting the printed surface of the printed matter, a shredder, or the like can be used. The filling device can receive information on the display range of the secret information on the printing surface from the control unit of the printing system, and can fill only the display range. In this way, instead of providing a dedicated filling device, it is also possible to return the printed matter after collation to the print processing unit 20 by using a paper conveyance path for double-sided printing, and the print processing unit 20 can fill the secret information portion. .

  Next, the configuration of the print quality inspection apparatus 50 will be described with reference to FIG. As shown in the figure, the print quality inspection apparatus 50 includes a fixed storage 510, a CPU (Central Processing Unit) 520, a memory 530, a scanner interface 540, and a communication control unit 550 that are communicably connected to each other via a bus 560. Composed. The fixed storage 510 is a non-volatile storage device such as a hard disk, a flash memory, or a read-only memory, and stores a verification processing program 512 that describes the procedure of the verification processing described above. The verification process described above is realized by loading the verification processing program 512 into the memory 530 and executing it by the CPU 520.

  The memory 530 is a main storage of the device 50, and stores a collation processing program 512 and various information generated during processing of the program. The stored information includes fixed image information 532 representing the fixed image 120, individual image information 534 representing the individual image 140, and inspection image information 536 representing the inspection image 100. The fixed image information 532 and the individual image information 534 are acquired from the control unit that controls the main body of the printing system via the communication control unit 550. The control unit of the printing system main body creates an individual image 140 for printing processing based on print instruction data sent from a host computer (not shown), and individual image information 534 representing the individual image 140 is controlled by the control unit. Provided to the print quality inspection apparatus 50. The standard image information 532 may be provided from the host computer to the print quality inspection apparatus 50 via the system control unit, but the preprinted paper before printing the individual images is printed on a series of standard documents. It may be obtained by reading with the reading scanner 30 (or the scanner for the function when the printing system has a copying machine function) before starting the operation. Further, the inspection image information 536 is acquired from the reading scanner 30 via the scanner interface 540. In addition, the projection waveforms 210, 220, and 240 and the differential waveform 230 are also stored in the memory 530 and provided for processing.

  Next, a processing procedure of the print quality inspection apparatus 50 will be described with reference to FIG.

  This procedure is started when the print quality inspection apparatus 50 acquires the result of reading one printed matter from the reading scanner 30, that is, the inspection image 100 (S10). Before the start of this processing procedure, the fixed image information 532 and the individual image information 534 are acquired, and the projection waveforms 210 and 240 along one or more predetermined directions of the fixed image 120 and the individual image 140 are also calculated. It shall be.

  When the inspection image 100 is acquired, the print quality inspection device 50 calculates a projection waveform 220 along one or more predetermined directions of the inspection image 100 (S12). Then, by comparing the projection waveform 220 with the projection waveform 210 of the standard image 120 for each direction, how much displacement and enlargement / reduction occurs along each direction when the standard image 120 is printed. The conversion parameter is calculated (S14). This conversion parameter includes direction information in the case of positional deviation, and is indicated by a magnification (in the case of reduction, 1 or less) in the case of enlargement / reduction. Next, the print quality inspection apparatus 50 determines whether or not the obtained conversion parameters (for example, the positional deviation amount and the enlargement / reduction ratio) are within an allowable range registered in advance in the apparatus 50 (S16). ). In the determination of S16, for example, if any one of the conversion parameters for each direction of the calculated projection waveform deviates from the allowable range, the determination result is negative (N). Of course, the conversion parameters in these directions may be comprehensively determined. If it is determined in this determination that the conversion parameter is not within the allowable range, the print quality inspection apparatus 50 determines that printing of the standard image 120 is defective (S18), and outputs the determination result to the control unit of the printing system. (S34). The control unit of the printing system processes the inspected printed matter according to the determination result.

  If it is determined in S16 that the conversion parameter is within the allowable range, the print quality inspection apparatus 50 next geometrically converts the projection waveform 210 in each direction of the standard image 120 by the conversion parameter in the direction. By performing (for example, translation and enlargement / reduction), the projection waveform 220 in each direction of the inspection image 100 is matched with the position and scale. Then, for each direction, the difference waveform 230 is formed by subtracting the projection waveform 210, which is a combination of position and scale, from the projection waveform 220 of the inspection image 100 (S20). This is because the fixed image 120 is aligned and scaled with a portion corresponding to the fixed image 120 in the inspection image 100, and the fixed image having the position and the scale is subtracted from the inspection image 100, and the difference obtained thereby. This is equivalent to the process for obtaining the projection waveform in each direction for the image.

  When the difference waveform 230 in each direction is obtained, the print quality inspection apparatus 50 next compares the difference waveform 230 and the projection waveform of the individual image 140 for each direction, thereby each of the individual images 140 when the individual image 140 is printed. Directional geometric conversion parameters, such as misalignment and enlargement / reduction amount are calculated (S22). Next, the print quality inspection apparatus 50 determines whether or not the obtained conversion parameter is within an allowable range registered in the apparatus 50 (S24). The allowable range in this determination can be determined independently of the allowable range in the determination of S16. If it is determined in this determination that the conversion parameter is not within the allowable range, the print quality inspection apparatus 50 determines that the printing of the individual image 140 is defective (S26), and outputs the determination result to the control unit of the printing system. (S34). The control unit of the printing system processes the inspected printed matter according to the determination result.

  If it is determined in S24 that the conversion parameter is within the allowable range, the print quality inspection apparatus 50 geometrically converts the fixed image 120 by the conversion parameter in each direction obtained in S14, and the individual image 140 in S22. The resultant image is geometrically converted by the calculated conversion parameter in each direction, and the two images are combined to create a combined image 160 (S28). Then, the print quality inspection apparatus 50 collates the inspection image 100 by pattern matching or the like using the composite image 160 as a template (S30), and if the print quality is good or defective based on the result of the collation. Determines the location of the defect (S32). The determination result is supplied to the control unit of the printing system (S34), and the control unit processes the inspected printed matter according to the determination result.

  The embodiment described above is merely an example.

For example, in the above embodiment, when the differential waveform 230 is obtained, the projection waveform 210 of the standard image 120 is geometrically transformed so as to match the projection waveform 220 of the inspection image 100 (hereinafter referred to as transformation T for identification). In contrast to this, the inverse of the projection waveform 220 is subtracted from the projection waveform 220, but the inverse transformation T ^-1 of T is applied so that the projection waveform 220 of the inspection image 100 matches the projection waveform 210 of the standard image 120, and the former is subtracted from the latter. A differential waveform (hereinafter referred to as differential waveform 230 ′ for identification) can be obtained. The difference waveform 230 can be obtained by performing the conversion T on the difference waveform 230 ′, and thereafter, the processing after S22 described above may be executed.

Further, after obtaining the above-described difference waveform 230 ′, an inverse transformation T ⁻¹ is applied to the projection waveform 240 of the individual image 140, and the resulting projection waveform 240 ′ and the difference waveform 230 ′ are It is also possible to calculate geometric transformation parameters for aligning and scaling the former to the latter. Using the geometric transformation parameters obtained in this way, the processes after S24 may be executed.

  In the above description, the case where the printed matter in which the individual image 140 is printed on the preprinted paper on which the standard image 120 is printed in advance has been described. However, as is clear from the processing content described above, the present invention is not limited to this. The present invention is not limited to such a case, and the present invention can be applied to the inspection of a printed matter created by printing a plurality of images on the same sheet surface by a plurality of printing processes.

  Further, after obtaining the geometric transformation for matching the individual image to the inspection image, the geometric transformation for fitting the fixed image to the inspection image may be obtained.

  Furthermore, conversely, after obtaining a geometric transformation for fitting the inspection image to the individual image, a geometric transformation for fitting the examination image to the individual image may be obtained.

  In the above description, the processing for a black and white binary image has been described as an example. However, the method of the above embodiment can be applied even when the inspection image 100, the fixed image 120, and the individual image 140 are color images. In this case, the color images may be binarized, and the geometric transformation parameters of the fixed image 120 and the individual image 140 may be calculated by the above-described method based on the projection waveforms of the binarized images. Then, the color standard image 120 and the individual image 140 are geometrically converted using the calculated conversion parameters, a composite image 160 is created, and this composite image is collated with the color inspection image 100.

  In addition to the above, various modifications are possible within the scope of the technical idea of the present invention shown in the claims.

It is a figure which shows the relationship between a fixed form image and a test | inspection image in the specific example of the printed matter used for description for the processing content of this invention. It is a figure which shows the relationship between an individual image and a test | inspection image in the specific example of the printed matter used for description for the processing content of this invention. It is a figure for demonstrating the projection waveform of a fixed form image. It is a figure for demonstrating the projection waveform of a test | inspection image. It is a figure which shows the comparison process of the projection waveforms between a fixed form image and a test | inspection image. It is a figure which shows the difference waveform of the projection waveforms between a fixed form image and a test | inspection image. It is a figure for demonstrating the projection waveform of an individual image. It is a figure for demonstrating the process which produces | generates a synthesized image from a fixed form image and an individual image. It is a figure which shows typically one projection waveform compared. It is a figure which shows typically the other projection waveform compared. It is explanatory drawing showing the local peak detection method for detecting the mode of the general unevenness | corrugation of a projection waveform. It is explanatory drawing which shows the operation | movement principle of the means which classify | selects a local peak. It is explanatory drawing showing the process for detecting the width | variety of a projection waveform. It is explanatory drawing showing the function for calculating | requiring the width | variety of a projection waveform. It is explanatory drawing showing an example of a peak list. It is explanatory drawing showing the outline | summary of the procedure which estimates a conversion parameter. It is explanatory drawing showing the outline | summary of the procedure which estimates a conversion parameter. It is explanatory drawing showing an example of a restriction | limiting of a voting area. 1 is a diagram illustrating a schematic configuration of an example of a printing system to which a print quality inspection apparatus according to the present invention is applied. It is a functional block diagram which shows an example of the hardware constitutions of the print quality inspection apparatus which concerns on this invention. It is a flowchart which shows an example of the process which the printing quality inspection apparatus which concerns on this invention performs.

Explanation of symbols

  100 inspection image, 120 fixed image, 140 individual image, 160 composite image, 210, 220, 240 projected waveform, 212, 222, 232, 242 characteristic peak, 230 differential waveform.

Claims (21)

A print quality inspection apparatus for inspecting a printed matter in which a first reference image and a second reference image are printed by separate printing processes on the same sheet surface,
Calculate a first geometric transformation between the first reference image and a first corresponding image portion corresponding to the first reference image in the printed surface image obtained by reading the printed surface of the printed matter. A first conversion calculation unit,
A second transformation calculation unit for calculating a second geometric transformation between the second reference image and a second corresponding image portion corresponding to the second reference image in the printing surface image;
A first determination unit that determines the quality of the print surface image using the first reference image, the second reference image, the first geometric transformation, and the second geometric transformation;
A print quality inspection apparatus comprising:
The print quality inspection apparatus according to claim 1,
The print quality inspection apparatus, wherein the first conversion calculation unit calculates the first geometric conversion by comparing predetermined image features of the print surface image and the first reference image. . The print quality inspection apparatus according to claim 2,
The second conversion calculation unit includes a difference between the predetermined geometric feature of the first reference image and the predetermined geometric feature of the print surface image, and the difference between the predetermined geometric feature of the printed surface image, A print quality inspection apparatus characterized in that the second geometric transformation is calculated by comparing the predetermined image feature of two reference images. The print quality inspection apparatus according to claim 2,
The second conversion calculation unit is configured such that a difference between an image obtained by reflecting an inverse transformation of the first geometric transformation on the predetermined image feature of the print surface image and the predetermined image feature of the first reference image. And the predetermined image feature of the second reference image to calculate the second geometric transformation. The print quality inspection apparatus according to any one of claims 2 to 4,
The print quality inspection apparatus, wherein the predetermined image feature is a projection waveform obtained by projecting an image in a predetermined direction within a plane of the image. The print quality inspection apparatus according to any one of claims 2 to 4,
The print quality inspection apparatus, wherein the predetermined image feature is a plurality of projection waveforms obtained by projecting an image in a plurality of predetermined directions in a plane of the image. The print quality inspection apparatus according to claim 6,
The first conversion calculation unit includes a plurality of feature points that satisfy a predetermined determination condition in the projection waveform of the first reference image, and the predetermined determination condition in the projection waveform in the print surface image. A geometric transformation that best matches a plurality of feature points satisfying the above is calculated using a Hough transform operation, and the geometric transformation obtained thereby is defined as the first geometric transformation. A print quality inspection device. The print quality inspection apparatus according to claim 6,
The second conversion calculation unit obtains a difference waveform between the projection waveform of the first reference image and the projection waveform of the print surface image, which is obtained by reflecting the first geometric transformation on the projection waveform of the first reference image. And a geometric transformation that best matches the plurality of feature points satisfying the predetermined determination condition and the plurality of feature points satisfying the predetermined determination condition in the projected waveform of the second reference image. A print quality inspection apparatus characterized in that a geometric transformation calculated by using a Hough transform operation is used as the second geometric transformation. The print quality inspection apparatus according to claim 6,
The second conversion calculation unit obtains a difference waveform between the projection waveform of the print surface image reflecting the inverse transformation of the first geometric transformation and the projection waveform of the first reference image. A geometrical feature that best matches a plurality of feature points satisfying the predetermined determination condition in the waveform and a plurality of feature points satisfying the predetermined determination condition in the projection waveform of the second reference image. A print quality inspection apparatus characterized in that a transformation is calculated using a Hough transform operation, and the geometric transformation obtained thereby is the second geometric transformation. The print quality inspection apparatus according to claim 1,
The print quality inspection apparatus according to claim 1, wherein the first reference image is an image including a ruled line, a table ruled line, or a frame line. The print quality inspection apparatus according to claim 1,
The first determination unit converts the first post-conversion image obtained by applying the first geometric transformation to the first reference image, and the second geometric transformation to the second reference image. Forming a composite image by combining the second converted image obtained by applying, and determining the quality of the print surface image based on a comparison between the print surface image and the composite image,
A print quality inspection apparatus characterized by that. The print quality inspection apparatus according to claim 1,
A second determination unit configured to determine that the print surface image is defective when the first geometric conversion calculated by the first conversion calculation unit deviates from a predetermined allowable range; 2. The print quality evaluation apparatus according to claim 1, wherein the determination process by the first determination unit is not executed when the determination unit determines that the print surface image is defective. The print quality inspection apparatus according to claim 1,
A third determination unit configured to determine that the print surface image is defective when the second geometric conversion calculated by the second conversion calculation unit deviates from a predetermined allowable range; 3. The print quality evaluation apparatus according to claim 1, wherein the determination process by the first determination unit is not executed when the determination unit determines that the print surface image is defective. A print quality inspection method executed by a computer system for inspecting a printed matter in which a first reference image and a second reference image are printed by separate printing processes on the same sheet surface,
Calculate a first geometric transformation between the first reference image and a first corresponding image portion corresponding to the first reference image in the printed surface image obtained by reading the printed surface of the printed matter. And
Calculating a second geometric transformation between the second reference image and a second corresponding image portion corresponding to the second reference image in the printing surface image;
Using the first reference image, the second reference image, the first geometric transformation, and the second geometric transformation to determine the quality of the printed surface image;
Method. 15. The method of claim 14, wherein
In the step of calculating the first geometric transformation, the first geometric transformation is calculated by comparing predetermined image features of the printing surface image and the first reference image. Feature method. The method of claim 15, comprising:
In the step of calculating the second geometric transformation, the predetermined image feature of the first reference image reflects the first geometric transformation and the predetermined image feature of the print surface image And calculating the second geometric transformation by comparing the difference between the second reference image and the predetermined image feature of the second reference image. The method of claim 15, comprising:
In the step of calculating the second geometric transformation, the predetermined image feature of the print surface image reflects an inverse transformation of the first geometric transformation and the predetermined reference image of the first reference image. Calculating the second geometric transformation by comparing the difference between the second image and the predetermined image feature of the second reference image. A program for causing a computer system to function as a print quality inspection device for inspecting a printed matter in which a first reference image and a second reference image are printed by separate printing processes on the same sheet surface, Computer system
Calculate a first geometric transformation between the first reference image and a first corresponding image portion corresponding to the first reference image in the printed surface image obtained by reading the printed surface of the printed matter. A first conversion calculation unit,
A second transformation calculation unit for calculating a second geometric transformation between the second reference image and a second corresponding image portion corresponding to the second reference image in the printing surface image;
A first determination unit that determines the quality of the print surface image using the first reference image, the second reference image, the first geometric transformation, and the second geometric transformation;
Program to make it function. The program according to claim 18, wherein
The first conversion calculation unit calculates the first geometric conversion by comparing predetermined image features of the print surface image and the first reference image. The program according to claim 19, wherein
The second conversion calculation unit includes a difference between the predetermined geometric feature of the first reference image and the predetermined geometric feature of the print surface image, and the difference between the predetermined geometric feature of the printed surface image, A program characterized in that the second geometric transformation is calculated by comparing the predetermined image feature of two reference images. The program according to claim 19, wherein
The second conversion calculation unit is configured such that a difference between an image obtained by reflecting an inverse transformation of the first geometric transformation on the predetermined image feature of the print surface image and the predetermined image feature of the first reference image. And calculating the second geometric transformation by comparing the predetermined image feature of the second reference image with the predetermined image feature.

Images