JP2022132991A - Image processing device, control method thereof, and program - Google Patents

Abstract

To overcome a disadvantage in which printed result inspection processing takes a very long time and results in a decrease in productivity of a printing apparatus when alignment based on nonlinear transformation is used in the inspection of printed results although the alignment accuracy is high.SOLUTION: Alignment parameters used in alignment based on nonlinear transformation that has already been performed are reused when the alignment based on the nonlinear transformation is performed to inspect print defects of printed sheets continuously output from a printing apparatus.SELECTED DRAWING: Figure 6

Description

The present invention relates to a technology for inspecting print results.

When forming an image on a recording medium with a printing apparatus, for example, coloring materials such as ink or toner may adhere to unintended locations, resulting in stains, or insufficient coloring materials may adhere to locations where an image should be formed. In some cases, color omission may occur in which the intended color cannot be reproduced. Print defects such as stains and missing colors reduce the quality of printed matter. Therefore, in applications such as so-called commercial printing, it is necessary to inspect whether there are any printing defects in the resulting product output from the printing apparatus before delivery. Visual inspection, in which an inspector visually inspects for printing defects, requires a lot of time and cost, so in recent years, an inspection system that automatically performs inspection without relying on visual inspection has been proposed.

As an automatic inspection method, there is a method of reading a printed matter output from a printing apparatus with a scanner and comparing the obtained read image with a reference image (correct image) to detect printing defects. In the case of this method, alignment at the time of comparing images greatly affects inspection accuracy, so it is important to perform alignment with high accuracy. As a general alignment method, a method of extracting feature points and then performing linear transformation such as projective transformation to align the positions of images is known. Japanese Patent Application Laid-Open No. 2002-200001 discloses a technique for speeding up processing while maintaining inspection accuracy by extracting feature points at the leading edge and trailing edge of a read image in the transport direction and aligning them by affine transformation. ing. However, the technique of linearly transforming the entire image, such as that disclosed in Japanese Patent Application Laid-Open No. 2002-200310, has a theoretical limit in that it cannot deal with local positional deviations due to unevenness in transportation and stretching of paper. In this regard, as a more accurate alignment method, a method of locally matching images by nonlinear transformation using control points arranged at regular intervals (for example, FFD (Free-Form Deformations)) is known.

JP 2020-118497 A

With the above-described alignment method using nonlinear transformation, it is possible to deal with local misalignment. However, this alignment method using nonlinear transformation requires a very long time for processing such as updating control points and calculating pixel positions after alignment. Therefore, there is a problem that the inspection takes a long time, and as a result, the productivity of the printing apparatus is lowered.

An image processing apparatus according to the present disclosure is an image processing apparatus that inspects a printed sheet output from a printing apparatus, and includes an acquisition unit that acquires a read image by reading the printed sheet, and an image that is acquired by the acquisition unit. positioning means for aligning the read image with a reference image serving as a reference in the inspection; and detecting the presence or absence of print defects based on the difference between the read image after the positioning and the reference image. and an inspection means, wherein the alignment means performs nonlinear alignment by performing nonlinear transformation on the read images of each of a plurality of continuously output printed sheets to align them with the reference image. It is characterized by reusing the parameters used in the performed nonlinear alignment.

According to the technique of the present disclosure, high-speed processing can be performed under a registration technique based on nonlinear transformation. As a result, high-precision inspection can be performed without reducing the productivity of the printing apparatus.

FIG. 2 is a diagram showing a configuration example of an entire printing system; 2 is a block diagram showing the functional configuration of an image processing apparatus; FIG. 4 is a flowchart showing a rough flow of inspection processing performed by the image processing apparatus; 4 is a flowchart showing details of defect detection processing; 4A and 4B are diagrams showing an example of the shape of an enhancement filter; FIG. 6 is a flowchart showing details of alignment processing according to the first embodiment; 4A to 4C are diagrams for explaining alignment processing; FIG. 4A to 4C are diagrams for explaining alignment processing; FIG. 4A and 4B are views for explaining alignment processing; FIG. FIG. 6 is a diagram showing an example of a UI screen for selecting an operation mode during alignment; The figure which shows an example of UI screen which notifies a test result. 4A and 4B are diagrams for explaining sheet misalignment; FIG. 9 is a flowchart showing details of alignment processing according to the second embodiment;

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. In addition, the following embodiments do not limit the invention according to the scope of claims. Although multiple features are described in the embodiments, not all of these multiple features are essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant description is omitted.

[Embodiment 1]
It is empirically known that local misalignment caused by uneven transport or stretched paper usually does not change between successive printing processes. Therefore, in this embodiment, when inspecting printed materials continuously while performing alignment by nonlinear transformation, the time required for alignment processing is reduced by reusing the alignment parameters used in the inspection of the previous printed material. will be shortened to speed up the inspection. In this specification, alignment by linear transformation is called "linear alignment", and alignment by nonlinear transformation is called "nonlinear alignment".

<Configuration of printing system>
First, referring to FIG. 1, the configuration of the entire printing system according to the present embodiment will be described. The printing system shown in FIG. 1 has an image processing apparatus 100 , a printing server 180 , and a printing apparatus 190 . The print server 180 has a function of generating a print job of a document to be printed and inputting the print job to the printing device 190 . A plurality of external devices (not shown) may be communicatively connected to the print server 180 via a network, and requests for generating print jobs, print data, and the like may be received from these external devices.

The printing device 190 has a function of forming an image on a recording medium (paper) based on a print job input from the print server 180 . As a printing method in the printing device 190, a known printing method such as an offset printing method, an electrophotographic method, or an inkjet method can be applied. In this embodiment, the case where the electrophotographic method is adopted will be described as an example, but it goes without saying that the present invention is not limited to this. The printing apparatus 190 has a paper feeding section 191, and the user sets printing paper in the paper feeding section 191 in advance. When a print job is input, the printing apparatus 190 forms an image specified by the print job on the surface or both sides of the paper set in the paper feed unit 191 while conveying the paper along the conveying path 192 to form the image. The printed sheet (hereinafter referred to as a “printed sheet”) is conveyed to the image processing apparatus 100 .

The image processing apparatus 100 performs an inspection process for checking the presence or absence of print defects on the printed sheet conveyed from the printing apparatus 190 . In other words, the image processing apparatus 100 functions as an inspection processing apparatus for printed matter. The image processing apparatus 100 includes a CPU 101, a RAM 102, a ROM 103, a main storage device 104, and an image reading device 105 inside. The image processing apparatus 100 also includes an interface (I/F) 106 with the printing apparatus 190 , a general-purpose interface (I/F) 107 , a user interface (UI) panel 108 , and a main bus 109 . Further, the image processing apparatus 100 includes a transport path 110 connected to the transport path 192 of the printing apparatus 190, an output tray 111 for printed sheets that have passed the inspection process, and an output tray 111 for printed sheets that have failed the inspection process. A tray 112 is provided. In the present embodiment, the output trays are divided into two types, i.e., pass and fail inspection processing.

<Hardware Configuration of Image Processing Apparatus>
A CPU 101 is a processor that controls the entire image processing apparatus 100 . The RAM 102 functions as a main memory, work area, etc. of the CPU 101 . The ROM 103 stores programs executed by the CPU 101 . The main storage device 104 stores applications executed by the CPU 101, data used for image processing, and the like. The image reading device (scanner) 105 reads one side or both sides of the printed sheet conveyed from the printing device 190 on the conveying path 110 and obtains image data. Specifically, the image reading device 105 uses one or more reading sensors (not shown) provided near the conveying path 110 to optically read one side or both sides of the conveyed printed sheet. The reading sensor may be provided only on one side, or may be provided on both the front side and the back side. In the case where it is provided only on one side, after reading one side, the printed sheet is turned upside down using a double-sided conveying path (not shown) in the conveying path 110, and the reading sensor reads the other side again. good.

The printing device I/F 106 is an interface for synchronizing the timing of conveying printed sheets with the printing device 190 and communicating the operating status of each other. A general-purpose I/F 107 is a serial bus interface such as USB or IEEE 1394, through which a user brings out data such as a log or imports some data into the image processing apparatus 100 . The UI panel 108 is, for example, a liquid crystal display (display unit), functions as a user interface of the image processing apparatus 100, displays the current status and settings, and informs the user. Alternatively, a touch panel type liquid crystal display may be used, and a user may operate a displayed button to receive an instruction from the user.

A main bus 109 is a bus that connects each unit of the image processing apparatus 100 and exchanges data and the like. In accordance with instructions from the CPU 101 through the main bus 109, each part of the image processing apparatus 100 and various internal parts of the printing system can be operated. For example, moving the transport path 110 in synchronization with the transport path 192 of the printing apparatus 190, or switching between the output tray 111 and the output tray 112 as the discharge destination of the printed sheet according to the inspection result. can be done. Also, an image processing device. 00 may include a GPU specialized for image processing in addition to the CPU 101 .

The image processing apparatus 100 according to the present embodiment causes the image reading apparatus 105 installed on the conveying path 110 to read the printed sheet conveyed from the printing apparatus 190, and the obtained read image is processed as follows. Perform the inspection process to be described. As a result of the inspection process, if no printing defect is detected from the printed sheet, it is determined that the inspection has passed and the printed sheet is discharged to the output tray 111. If any printing defect is detected, it is determined that the inspection has failed and the printed sheet has been printed. The sheet is discharged to output tray 112 . In this way, only the products that have passed the inspection and whose quality has been guaranteed can be collected in the output tray 111 as deliverables.

<Outline of inspection process>
Next, an overview of inspection processing executed by the image processing apparatus 100 will be described with reference to the block diagram of FIG. 2 and the flowchart of FIG. As shown in FIG. 2, the image processing apparatus 100 includes an image acquisition unit 201, an inspection item setting unit 202, an alignment parameter storage unit 203, an alignment processing unit 204, an inspection parameter determination unit 205, an inspection processing unit 206, an inspection result It has each processing unit of the output unit 207 . Then, for example, the CPU 101 expands a program corresponding to the flowchart of FIG. 3 stored in the ROM 103 into the RAM 102 and executes the program, thereby realizing the functions of the above processing units. In the following description, symbol "S" means step.

In S301, the inspection item setting unit 202 sets one or more inspection items to be performed in inspection processing. Along with setting the inspection items, the inspection parameter setting unit 205 sets inspection parameters corresponding to one or more set inspection items. Prior to setting the inspection items, the user selects a desired inspection item via an inspection item setting screen (not shown) displayed on the UI panel 108 . Inspection items include point-shaped defects, line-shaped (streak) defects, image unevenness, surface shape distortion, etc. The user can select any inspection item from options corresponding to the types of these print defects. do. It should be noted that, if no user selection is made, default examination items may be set. The inspection parameter setting unit 205 sets inspection parameters corresponding to the inspection items set by the inspection item setting unit 202 in accordance with the setting of the inspection items. The inspection parameters include a filter according to the set inspection item (type of print defect), a threshold value for determining the presence or absence of the print defect, and the like. The setting of inspection parameters will be described later.

Subsequently, in step S302, the image acquisition unit 201 acquires a reference image (hereinafter referred to as a "reference image") for determining whether or not the printed sheet has a print defect. (also called "image") is acquired from the RAM 102 or the main storage device 104 . Further, in S303, the image acquisition unit 201 causes the image reading device 105 to read the printed sheet conveyed from the printing device 190, and obtains a read image to be inspected (hereinafter referred to as an "inspection target image"). ). Note that the image to be inspected may be read in advance by the image reading device 105 and read image data held in the main storage device 104 may be acquired.

Next, in S304, the inspection item setting unit 202 sets an inspection item (initial value) to be executed first among the inspection items set in S301. This initial value may be determined in any order, such as selecting the inspection item first selected by the user as the initial value if there is no particular priority for each of the set inspection items. Then, in subsequent S305, the alignment parameter storage unit 203, the alignment processing unit 204, and the inspection processing unit 206 cooperate to perform defect detection on inspection items of interest as execution targets (hereinafter referred to as "notable inspection items"). Execute the process. Details of the defect detection process will be described later.

After completing the defect detection process for the inspection item of interest, in S306, the inspection processing unit 206 determines whether or not the defect detection process for all the inspection items set in S301 has been completed. If defect detection processing has been completed for all inspection items, the process proceeds to S308, and if there are unprocessed inspection items remaining, the process proceeds to S307. In S307, the inspection item setting unit 202 updates the inspection item of interest. That is, the next target inspection item is determined from the unprocessed inspection items, and the process returns to S305 to repeat the same processing.

In S<b>308 , the inspection result output unit 207 displays the inspection results of all inspection items set in S<b>301 on the UI panel 108 . The details of the inspection result display processing will be described later.

The outline of the inspection process in the image processing apparatus 100 has been described above. Such inspection processing is performed for each printed sheet conveyed from the printing apparatus 190 .

<Defect detection processing>
Next, the defect detection process (S305) described above will be described in detail with reference to the flowchart of FIG. In this embodiment, a method of detecting print defects from an inspection target image by comparing it with a reference image obtained by reading a print-processed sheet that has been confirmed to have no print defects in advance will be described.

First, in S401, the alignment processing unit 204 performs alignment processing for aligning the inspection target image with the reference image so that the reference image and the inspection target image can be easily compared. Details of the alignment process will be described later. Subsequently, in S402, the inspection processing unit 206 generates an image (hereinafter referred to as a "difference image") indicating the difference between the image to be inspected after alignment and the reference image. This difference image is obtained, for example, by comparing corresponding pixels in both images and obtaining a difference value between pixel values (eg, RGB values) for each pixel.

In S403, the inspection processing unit 206 executes filter processing for emphasizing a specific shape on the difference image generated in S402. What kind of filter is used in the filtering process is determined according to the inspection item of interest. For example, when the inspection item of interest is the detection of a point-like defect, a filter for emphasizing the point-like defect as shown in FIG. 5A is used to detect the line-like defect. In this case, a filter for emphasizing linear defects is used as shown in FIG. 5(b).

Next, in S404, the inspection processing unit 206 executes binarization processing on the filtered difference image. This binarization process sets the pixel value to "1" if the difference value of each pixel in the difference image after filtering is equal to or greater than the threshold value, and sets the pixel value to "0" if the difference value is equal to or less than the threshold value. As a result, each pixel is converted into a binary image represented by "1" or "0". Here, for example, assume that the filter size applied in the filtering process is reduced. In this case, point-like defects of smaller size are emphasized and easily detected. If the threshold value in the binarization process is made small, even if the difference value is small, it exceeds the threshold value and becomes "1", and is easily detected as a defective pixel. That is, it is possible to detect even defective pixels with a smaller contrast. In the above-described S301, the filter size, the threshold in the binarization process, and the like are set as inspection parameters.

Subsequently, in S405, the inspection processing unit 206 determines whether there are defective pixels (pixels whose pixel value is “1”) whose difference value exceeds the threshold based on the binary image obtained by the binarization process. determine whether or not If a defective pixel exists, the process proceeds to S406, and if not, the process exits.

In S406, the inspection processing unit 206 stores the position information (coordinate information) of the detected defective pixel in the RAM 102 in association with the inspection item of interest, and exits from this process.

The above is the details of the inspection process. Such inspection processing is repeated by the number of inspection items set in S301.

<Alignment processing>
Next, with reference to the flowchart of FIG. 6, the above-described alignment processing (S401) will be described in detail. Below, the image to be inspected I shown in FIG. 7A is aligned with the reference image T shown in FIG. 7B, and the image after alignment shown in FIG. The case of obtaining I' will be described as an example. In the following description, I(x, y), T(x, y), and I'(x, y) represent pixel values at coordinates (x, y).

In S<b>601 , the alignment processing unit 204 acquires the previous alignment parameters saved in the alignment parameter storage unit 203 . Here, the previous alignment parameters are the conversion formula (3) and the control point P _{l used in the previous inspection process (that is, the (N−1)th sheet with respect to the Nth sheet),} means the location information of _m . FIG. 8(a) shows a state in which control points P _{l and m} are arranged in a grid pattern at regular intervals in the image I to be inspected shown in FIG. 7(a). In the registration process, by updating the positions of the control points P _l,m arranged in the image I to be inspected, the image I to be inspected is non-linearly transformed into the post-alignment image I′.

In S602, the alignment processing unit 204 determines whether or not the previous alignment parameters could be acquired in S601. Immediately after starting the inspection process (when inspecting the first printed sheet), the previous alignment parameters are not stored in the alignment parameter storage unit 203 . Therefore, the previous alignment parameters can be acquired from the inspection process on the second and subsequent printed sheets. If the previous alignment parameters cannot be acquired (that is, if the first printed sheet is to be inspected), the process advances to step S603. On the other hand, if the previous registration parameters could be acquired (that is, if inspection is to be performed on the second and subsequent printed sheets), the process advances to step S605. Note that "previous time" basically means one page before (the N-1 page with respect to the Nth page), but is not necessarily limited to one page before.

In S603 when the previous alignment parameters could not be acquired, the alignment processing unit 204 performs initial alignment. Alignment by general linear transformation is applied to this initial alignment. For example, feature points are extracted, and projective transformation or affine transformation is performed so that the total sum of Euclidean distances of the extracted feature points is minimized.

In the next step S604, the alignment processing unit 204 arranges control points on the image I to be inspected. Specifically, as shown in FIG. 8A, L rows×M columns of control points P _1,m are arranged in a grid for the image I to be inspected. Note that L and M are positive integers. At this time, the distance δ between the control points is determined by the values of L and M and the size of the image to be inspected. For example, assume that the image I to be inspected is A3 (7016×9921 pixels) with a resolution of 600 dpi. In this case, with A3 size or larger (7050×9950 pixels) as a candidate, L=199, M=141, and the distance δ=50 pixels.

In S605, the alignment processing unit 204 updates the positions of the control points P _l,m arranged in the image I to be inspected. The control point to be updated here is the control point arranged after the initial alignment when the first printed sheet is to be inspected. Control points placed based on the previous alignment parameters obtained. The position of the updated control point P _l,m is obtained by the following equation (1).

In the above formula (1), μ represents a weighting factor, and has a value such as 0.1, for example. Also, μ may be changed according to the update speed of the control points P _l,m . ∇c is the pixel value of the inspection target image I′ and the pixel of the reference image T in the set D _l,m of pixels near the control point P _l,m (see FIGS. 8(b) and (c)) is a differential value of the sum of squares of the difference from the value, and is represented by the following equation (2).

In S<b>606 , the alignment processing unit 204 calculates the pixel position after alignment with the reference image T for each pixel of the image to be inspected I, and obtains the image after alignment I′. The pixel position after alignment in each pixel of the image to be inspected I is obtained by the following equation (3).

In the above equation (3), w(x, y) is the coordinates after alignment of the coordinates (x, y) in the image I to be inspected, and is represented by the following equation (4).

The bases B ₀ (t), B ₁ (t), B ₂ (t), and B ₃ (t) in the above formula (4) are respectively represented by the following formulas (5) to (8).

Also, as shown in FIG. 9, u=[x/δ]-1, v=[y/δ]-1, u′=x/δ-[x/δ], v′=y/δ-[ y/δ]. In the present embodiment, the grid points used to calculate the pixel positions in the aligned image I′ are p(u, v), p(u+1, v), . . . p(u+3, v+3). Although 16 points are used, the present invention is not limited to this. For example, four lattice points having a close Euclidean distance of (x, y) may be used.

In S<b>607 , the alignment processing unit 204 determines whether or not the alignment of the image to be inspected I has been completed and the alignment-aligned image I′ has been obtained. The determination as to whether or not the alignment is completed may be performed, for example, as follows. First, the distance d between the current position-aligned image I' and the reference image T is obtained using the following equation (9).

Then, determination may be made by performing threshold processing on the obtained distance d. Specifically, when the distance d obtained by the above formula (9) is equal to or less than the threshold, it can be determined that the alignment is completed. Now, assuming that the image to be inspected I is A3 (7016×9921 pixels) with a resolution of 600 dpi, X=7016 pixels and Y=9921 pixels. should be set to "5".

In S608, the alignment parameter storage unit 203 stores the alignment parameters at the time of completion of alignment for use in inspection of the next printed sheet. More specifically, the conversion formula of the above formula (3) for calculating the pixel position after alignment in each pixel of the inspection target image I and the arrangement information of the control points P _{l and m} updated in S605 are referred to as "previous are stored in the RAM 102 as "alignment parameters".

The above is the content of the alignment processing according to the present embodiment. For example, the user may be allowed to select either the non-linear alignment mode or the linear alignment mode via a UI screen for allowing the user to select the operation mode as shown in FIG. Then, the above processing may be performed when the nonlinear alignment mode is selected, and only normal linear alignment may be performed when the linear alignment mode is selected.

<Display of test results>
FIG. 11 is an example of a screen displayed on the UI panel in the inspection result display process (S308) described above. FIG. 11 is an example of a UI screen showing the inspection result when the inspection fails. The entire image to be inspected is displayed in the image display area 1101, and a character string indicating the type of the detected print defect is displayed. 1102 and 1103 are displayed. Furthermore, on the right side of the image display area 1101, position information (coordinate information) 1104 and 1105 indicating the position of the detected print defect is also displayed. Although not shown in the example of FIG. 11, there may be a page ID for specifying the image to be inspected, a button for switching pages to be displayed, and the like. Further, for example, a character string indicating the content may be displayed in a different color for each type of print defect.

<Modification>
In the above-described embodiment, when the "previous alignment parameters" are acquired, they are used, and the initial alignment (S603) and control point placement (S604) are skipped. However, even if the "previous alignment parameter" can be acquired, control may be performed so that the "previous alignment parameter" is not used in a predetermined case. Predetermined cases include cases where printing/inspection is restarted due to a paper jam occurring immediately before, or cases where printing conditions have changed (for example, the type of sheet is different), etc. This is the case when reuse of alignment parameters is not amenable. In these cases, initial alignment and placement of control points may be performed again.

As described above, in the present embodiment, in the printed sheet inspection process, the alignment parameters used in the most recent (N-1) alignment are reused in the next (Nth) alignment. . This makes it possible to shorten the time required for alignment processing.

[Embodiment 2]
In the first embodiment, it is assumed that unevenness in transportation and local misalignment during printing hardly change between successive printing processes. Next, a mode capable of coping with a case where the sheet is greatly misaligned between successive printing processes will be described as a second embodiment.

FIG. 12(a) shows a state in which the sheet is not misaligned during inspection, and FIG. 12(b) shows a state in which the sheet is misaligned during inspection. Suppose that the printed sheet to be inspected from now on is scanned with a large shift compared to the time of the previous inspection. In the present embodiment, assuming such a case, first, the obtained scanned image is subjected to projective transformation, and the image in a displaced state as shown in FIG. Convert to state image. Thereafter, non-linear alignment is performed using the alignment parameters used during the previous inspection. Note that the description of the contents common to the first embodiment, such as the configuration of the printing system, will be omitted, and the following will describe the alignment process, which is the point of difference.

<Alignment processing>
Next, the alignment processing (S401) according to this embodiment will be described in detail with reference to the flowchart of FIG.

First, in S1301, the alignment processing unit 204 performs initial alignment by general linear transformation, as in S603 of the flowchart of FIG. 6 according to the first embodiment. In subsequent S<b>1302 , the alignment processing unit 204 acquires the previous alignment parameters saved in the alignment parameter storage unit 203 . This process corresponds to S601 in the flowchart of FIG. 6 according to the first embodiment.

In S1303, the alignment processing unit 204 determines whether or not the previous alignment parameters could be acquired in S1302. Immediately after the inspection process starts, the "previous alignment parameters" are not stored in the alignment parameter storage unit 203. Therefore, the previous alignment parameters can be acquired only for the second and subsequent printed sheets. It is from the inspection process. If the previous registration parameters cannot be acquired (that is, if the first printed sheet is to be inspected), the process advances to step S1304. On the other hand, if the previous registration parameters could be acquired (that is, if inspection is to be performed on the second and subsequent printed sheets), the process advances to step S1305.

In the next step S1304, the alignment processing unit 204 arranges control points on the image I to be inspected. This process corresponds to S604 in the flowchart of FIG. 6 according to the first embodiment.

Each process of S1305 to S1308 corresponds to S605 to S608 in the flowchart of FIG. 6 according to the first embodiment, and since there is no particular difference, the description is omitted.
The above is the content of the alignment processing according to the present embodiment.

According to this embodiment, general linear alignment is performed at the beginning of the alignment process, and then non-linear alignment is performed using the alignment parameters used in the previous inspection process. As a result, it is possible to obtain the same effect as in the first embodiment while coping with the case where the printed sheet is scanned with a large deviation.

(Other examples)
The present invention supplies a program that implements one or more functions of the above-described embodiments to a system or device via a network or a storage medium, and one or more processors in the computer of the system or device reads and executes the program. It can also be realized by processing to It can also be implemented by a circuit (for example, ASIC) that implements one or more functions.

Claims (12)

An image processing device for inspecting a printed sheet output from a printing device,
acquisition means for reading the printed sheet and acquiring a read image;
alignment means for aligning the read image acquired by the acquisition means with a reference image serving as a reference in the inspection;
inspection means for detecting the presence or absence of print defects based on the difference between the read image after alignment and the reference image;
with
The positioning means is used in non-linear positioning that has already been performed when performing non-linear positioning by performing non-linear transformation on the read images of each of a plurality of printed sheets that are successively output and aligning them with the reference image. to reuse the parameters of the
An image processing apparatus characterized by: The positioning means reads the (N-1)th printed sheet when performing the non-linear positioning for the read image of the Nth printed sheet among the plurality of printed sheets that are continuously output. using the parameters used in said non-linear registration for the images;
2. The image processing apparatus according to claim 1, wherein: further comprising storage means for storing the parameters used when performing the non-linear alignment for the read image of the N-1 printed sheet;
The alignment means uses the parameters stored by the storage means to perform the non-linear alignment of the read image of the Nth printed sheet.
3. The image processing apparatus according to claim 2, wherein: 4. The image processing apparatus according to claim 3, wherein said alignment means does not perform said non-linear alignment using said parameters stored by said storage means in a predetermined case where said reuse is not suitable. . 5. The image processing apparatus according to claim 4, wherein the predetermined case is when a paper jam occurs and the inspection is restarted, or when printing conditions have changed. 6. The method according to any one of claims 1 to 5, wherein the parameter is arrangement information of control points arranged with respect to the read image acquired by the acquisition means for performing the nonlinear transformation. Image processing device. 6. The apparatus according to any one of claims 1 to 5, wherein the parameter is a conversion formula for calculating a post-alignment pixel position in each pixel of the read image acquired by the acquisition means. Image processing device. The alignment means are
linearly aligning the read image acquired by the acquisition means with the reference image by performing linear transformation as an initial alignment;
performing the non-linear alignment on the read image after the initial alignment;
8. The image processing apparatus according to any one of claims 1 to 7, characterized by: 9. The image processing apparatus according to claim 8, wherein the linear transformation is projective transformation or affine transformation. further comprising a user interface for allowing a user to select a first operation mode in which the linear alignment is performed and a second operation mode in which the nonlinear alignment is performed as the operation mode of the alignment means;
The alignment means performs the nonlinear alignment that reuses the parameters when the operation mode selected by the user interface is the second operation mode.
10. The image processing apparatus according to any one of claims 1 to 9, characterized by: A control method for an image processing device that inspects a printed sheet output from a printing device, comprising:
an obtaining step of reading the printed sheet to obtain a read image;
an alignment step of aligning the read image acquired in the acquisition step with a reference image serving as a reference in the inspection;
an inspection step of detecting the presence or absence of a print defect based on the difference between the read image after alignment and the reference image;
including
In the alignment step, when performing nonlinear alignment in which nonlinear transformation is performed on the read images of each of the plurality of printed sheets that are successively output to align them with the reference image, the nonlinear alignment is used in the already performed nonlinear alignment. to reuse the parameters of the
A control method characterized by: A program for causing a computer to function as the image processing apparatus according to any one of claims 1 to 10.

Images