JPH10289311A - Check method and device for printed matter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To check a printed matter even when its positioning is rough or the form is elastic when the printed matter image is compared with a sample image to detect the stains, the breakage, the wrong characters, etc., of the printed matter. SOLUTION: A check sample and the image of a printed matter to be checked are read via an image input part. Then a positioning pattern is selected at two points based on those read sample and image to calculate the position coordinates, and the address conversion is performed so as to put both images on each other (S1, S2). The edge components of both images are extracted, binarized and divided into blocks (S3 to S5). Then a position where both images put on each other with highest accuracy is decided by shifting the pixels one by one in every block, and the information on both images are compared with each other at the decided position to detects the different points between them (S6, S7).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inspection method for detecting dirt, chipping, and incorrect characters in printed matter. Especially,
This is an inspection method in which even when the position of an inspection print is roughly input, when comparing and detecting differences, differences in the quality control of the prints that do not result in waste paper are not detected as differences. .

[0002]

2. Description of the Related Art When printing a printed matter, a proofreading operation is performed. This is a task of checking the sample and the printed matter at the beginning of printing while comparing them over the entire surface. The emphasis here is on checking that there is no problem in the character part.
For example, in the field of package printing, a description of a product or the like is printed on a packaging material. In particular, if a medicine becomes dirty, chipped, or has a wrong character, a consumer may use the product incorrectly, which is extremely dangerous.

[0003] However, the inspection work imposes a heavy load on the operator due to the small size of the characters, the large number of multi-faced prints, and the like, and oversight rarely occurs.

[0004] In recent years, an apparatus for automatically inspecting a printing machine or a rewind inspection machine equipped with a camera and an image processing apparatus has appeared, but such an apparatus also performs 100% inspection at a high printing speed. Therefore, the resolution of the camera itself, which is an input device, is low, which is not suitable for such an inspection.

[0005] In view of this, an inspection apparatus has been developed which prepares a high-resolution image input apparatus such as a flatbed scanner, inputs a sample and a printed material to be inspected as an image, and performs image processing to detect a difference between the two. .

However, with such an apparatus, it is not possible to position the sample and the inspection product in exactly the same way and input an image. Then, the input sample image is displayed on a TV monitor or the like, and an appropriate pattern is selected in advance by the operator for alignment from the image, and the operator selects one of the images so that the sample image matches the inspection print image. The position must be adjusted for each sheet.

Here, the "pattern suitable for positioning" is a line drawing pattern in which the contrast between the ground color of the paper and the printing color is clear and there is no shape similar to the periphery. (Hereinafter, referred to as the alignment pattern)

When at least two corresponding position coordinates are obtained from the images of the sample and the inspection product, the expansion and contraction and inclination of the paper between the sample and the inspection product can be calculated. By affine-transforming any of the images of the sample and the test print using these values, it is possible to compare both images for each pixel.

However, even if such processing is performed, in many cases, the sample and the image after the inclination correction of the test print are not completely coincident with each other at the edge of the picture. This is due to a distortion when the local expansion and contraction of the printing paper is treated as uniform expansion and contraction such as affine transformation, or a case where the printing position is actually shifted.

As described above, even a portion which does not need to be detected as a difference is detected. Therefore,
It is necessary to take measures so that local differences that do not cause a problem in quality control of printed matter are not detected.

In particular, when a single color character is printed in a solid filled rectangle or the like, there is no problem at all even if the position of the character is slightly shifted.

In the case where the register is completely coincident on one side of the multi-faced pattern, the position of each color is shifted by several hundred μm on the other side. In most cases, it will not be defective.
For such a reason, the sample and the image after the inclination correction of the inspection printed matter often do not completely match at the edge of the picture.

However, since the fine character portion to be inspected most by this inspection device is composed of many edge portions,
If the inspection standard is set strictly, a portion having no problem will be detected as a difference. On the other hand, in order to prevent this, the inspection standard can only be set loosely. For example, if the character is “3” in the sample but the character is “8” in the inspection product, the character size is small. Has a problem that it is not detected.

[0014]

SUMMARY OF THE INVENTION The present invention has been proposed in view of such a situation, and a picture of a printed matter is taken in from a detection unit (image input unit) for each pixel, and the detected pixel is detected for each pixel. In the inspection method of comparing the picture information with a pre-input reference picture signal of each corresponding pixel and detecting a stain, a chip, a character difference, or the like of a printed matter, a simple inclination,
Inspection method and apparatus for printed matter that enables inspection even if positioning is rough or paper is expanded or contracted for the problem that precise inspection cannot be performed even if pixel unit comparison is performed after expansion / contraction correction processing Is provided.

[0015]

According to the present invention, a picture of a printed matter is fetched for each pixel from a detection section (image input section), and the detected picture information for each pixel is input to a corresponding pixel previously inputted. Dirt, chipping,
In the inspection method for detecting a character difference or the like, (1) a step of storing a pattern of a test sample serving as a reference in a reference memory as reference pattern information, and storing image information of a pattern of a test print in an image memory. (2) At least two patterns for positioning are selected from the reference pattern information to determine the position coordinates, the position coordinates of the pattern for positioning are determined from the image information of the pattern on the inspection print, and coordinate conversion is performed based on the position coordinates of both. Process. (3) A step of extracting the edge components of the picture of each of the test sample and the test print. (4) a step of converting the edge component of the picture into a binary image. (5) a step of dividing the binarized image into blocks each having a size (W pixels × H pixels). (6) A step of obtaining a position that most overlaps with the reference pattern information of the inspection sample while shifting the position of the image information of the pattern of the inspection printed matter by one pixel in each block unit. (7) A step of comparing the reference pattern information of the inspection sample and the image information of the inspection printed matter for each pixel at the position where the inspection is most overlapped, and detecting and displaying a difference between the two. And a printed matter inspection method.

(1) An image input device for storing a pattern of a test sample serving as a reference in a reference memory as reference pattern information, and storing image information of a pattern of a test print in an image memory.
(2) a coordinate conversion means which has a reference memory and an image memory for respectively storing the reference picture information and the picture information of the picture of the test print, selects at least two points of the picture for alignment, obtains the position coordinates, and performs coordinate transformation; Edge extracting means for extracting an edge component of each pattern;
A binarized image converting unit for converting into a binarized image; a block dividing unit for dividing the binarized image into blocks each having a size (W pixels × H pixels); An optimal position detecting means for obtaining a position where the reference pattern information of the test sample overlaps with the reference pattern information of the test sample while shifting the position by one pixel; and comparing the reference pattern information of the test sample with the image information of the test print at each position where the pattern overlaps most,
An image processing apparatus having a difference detecting means for detecting a difference between the two, and (3) an image for inputting and displaying an instruction for executing the above processing, displaying the difference between the two, and displaying an inspection result And a display device for inspecting printed matter.

[0017]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a processing flow of the inspection method of the present invention, and FIG. 2 shows a hardware configuration of the present embodiment. An image processing device (1) composed of a CPU unit, an image memory unit, an image input unit, an image display unit, etc., an image input device (2), a TV monitor for image display (3), a keyboard (4), a mouse (5) ) Is used. In the present embodiment, a case will be described in which main processing is performed by software using a flatbed scanner for the image input device (2).

The process of detecting the difference between the sample and the test print is basically performed by a method of comparing corresponding pixels of each image. However, as shown in FIG. 3 (a), it is difficult to accurately place the sample or the test print horizontally on the flatbed scanner. Gets worse. For this reason, simply comparing pixels at the same coordinates in an input image results in comparisons between different positions in the image, and is detected as a difference over the entire surface.

Further, even when the inclination is zero, there is a case where they do not completely match due to the expansion and contraction of the paper. Therefore, a process of correcting the image of the inspection print to the same angle and size as the sample image is performed. Now, if the corrected image and the sample image can be compared to each other to detect only the difference by pixel-by-pixel comparison, it is actually a simple expansion and contraction of the printed matter, a slight displacement of the printing, etc. In the comparison for each pixel, a difference is detected at an edge portion of the image. However, even if such differences do exist in the printed matter, they do not matter and must be ignored by the inspection device.

Generally, as a method for ignoring differences occurring in an edge portion due to a positional shift or the like, a method of masking an edge portion or a maximum value filter for converting a maximum value in an area centered on a pixel of interest as a new value. And a method of using a minimum value filter for converting a minimum value in a region centered on a pixel of interest as a new value to make the inspection sensitivity of an edge portion weaker.

However, when such a process is applied to a character portion, which is an important object of the present inspection apparatus, the character portion has a very large edge occupation ratio, so that a precise inspection capable of detecting a slight difference is possible. Can not expect.

Therefore, a process as shown in FIG. 1 is performed. First, a sample image and an image of a test print are read from the image input device, and the sample image is stored in the reference memory, and the image of the test print is stored in the image memory (processing step S1).

At least two patterns for positioning are selected from the reference pattern information of the sample image, the center of gravity of the pattern for positioning is obtained, and the center of gravity of the pattern for positioning is determined from the image information of the pattern on the inspection print. Then, based on the position coordinates of both, the image information address of the picture of the test print and the reference picture information address of the sample image are coordinate-transformed, and a process is performed in which both images overlap (processing step S2).
By performing coordinate conversion, inclination correction and expansion / contraction correction are performed.

A differential filter process such as a Sobel filter is applied to each of the image information of the pattern of the inspection print after the inclination and expansion / contraction correction processing and the reference pattern information of the sample image to extract an edge component of the pattern (processing step S3). .

Then, the edge component is binarized and converted into a binarized image (processing step S4).

Next, in order to cope with a local displacement, FIG.
As shown in (5), the target sample image and the image of the test print after the inclination and expansion / contraction correction processing are divided into blocks each having a pixel size (W pixels × H pixels) (processing step S5).
The pixel size for forming a block is arbitrarily determined depending on the resolution of the image input device and how large the inspection should be. The resolution of the image input device is 360 dpi
In the case of (14 pixels / mm), when inspecting a character of up to 10 points (3.5 mm), it is preferable to divide the block into 50 pixels × 50 pixels because 14 × 3.5 = 49.

Then, while shifting the position of the block of the edge binarized image of the inspection print material for each block, a process of obtaining the position that overlaps the sample edge binary image is performed (processing step S6).

Here, a block having a pattern of the character "A" in FIG. 4 will be described as an example. Actually, the same processing as this block is performed for all blocks. Consider an area around the block, which is several pixels larger than the block size, as indicated by a bold line in FIG. The size is (W + 2) × (H +
2) to (W + 6) × (H + 6).

As an explanation of the concept of a method of obtaining the most overlapping position by shifting one pixel at a time, (W + 2) × (H +
Perform in the case of 2). An edge binarized block image of the inspection print shown in FIG. 5A and one each for the left, right, upper, and lower sides.
Assume that the corresponding pixel is superimposed on the edge binarized block image of the sample image at the corresponding position shown in FIG.

In the block of W × H pixels in FIG. 5A, the upper left corner is (0, 0), and the lower right corner is (W, H) coordinates. Then, in the block of (W + 2) × (H + 2) in FIG. 5C, the upper left can be expressed as (−1, −1) and the lower right can be expressed as (W + 1, H + 1). Here, when the upper left (0, 0) of FIG. 5A is shifted and superimposed on the upper left (-1, -1) of FIG. 5C, the result becomes as shown in FIG. When XOR (exclusive OR) processing is performed on all the pixels in H), the result is as shown in FIG. 5F, and the number of pixels that have become 1 after the processing is counted.

The pixel which becomes 1 after the XOR processing shown in FIG. 5 (f) indicates a state where the pixels of FIG. 5 (a) and FIG. 5 (c) do not match, and the pixel which becomes 0 shows a state where it matches. Will be displayed. That is, the number of pixels having become 1 is counted, and the smaller the number is, the higher the degree of coincidence between the two patterns is. If they completely match, the number of pixels that become 1 becomes zero.

Next, the upper left (0, 0) of FIG.
With respect to (c), three pixels are moved in the X (horizontal) direction and three pixels are moved in the Y (vertical) direction, one pixel at a time. That is, the upper left (0, 0) of FIG.
-1), (1, -1), (-1, 0), (0, 0),
(1, 0), (-1, 1), (0, 1), and (1, 1) are sequentially moved and overlapped. For these nine positions,
XOR processing is performed, and the number of pixels that become 1 is counted. Here, the position where the number of pixels having become 1 is the smallest can be determined as the position where the overlap between the sample image and the image of the inspection printed matter is the best.

Once the position of the best overlap in the block is determined, the sample image and the inspection image are compared pixel by pixel at that position to detect the difference between the two. By performing such processing, it is possible to minimize the difference detection at the pattern edge portion due to poor alignment.

Incidentally, the processing for obtaining the position with the best overlap while shifting the pattern in the block may be a method using a correlation coefficient, but the calculation amount becomes large. On the other hand, according to the present method, an overlap state can be determined only by XOR processing after first converting to an edge binarized image, so that high-speed processing is possible. This method is suitable because the influence of the displacement is remarkable at the edge portion.

Incidentally, the process of comparing the sample and the inspected product on a pixel-by-pixel basis after performing the position alignment for each block to find the difference between the two is performed by a generally used image processing method.

[0036]

According to the present invention, a picture of a printed matter is fetched for each pixel from a detecting section (image input section), and the detected picture information of each pixel is compared with a pre-input reference picture signal of each corresponding pixel to print the picture. By using the present invention, in an inspection device that detects dirt, chipping, character difference, and the like, the image after the inclination and expansion / contraction correction processing is divided into blocks, and the position is shifted in units of blocks, so that the position having the best overlap is obtained. Since the sample image and the inspection print image can be compared for each pixel, precise inspection becomes possible, and it becomes possible to detect partial chipping or dirt of fine characters, which has been difficult in the past.

[Brief description of the drawings]

FIG. 1 is a flowchart showing an inspection method of the present invention.

FIG. 2 is a diagram showing a hardware configuration of an embodiment of the present invention.

FIG. 3A is a diagram illustrating a state in which a sample or an inspection print is accurately and horizontally placed on a flatbed scanner. (B)
FIG. 3 is a diagram showing a state in which a sample or an inspection print is placed obliquely on a flatbed scanner.

FIG. 4 is a diagram illustrating a state in which a block is divided.

FIG. 5A is a block image showing a state in which a character “A” is composed of a block size (W × H). (B)
FIG. 3A is a binarized block diagram of FIG. (C) is a block image showing a state in which the character “A” is composed of a block size (W + 2) × (H + 2). (D) is a binarized block diagram of (c). (E) is a diagram showing a state in which (a) and (c) are superimposed. (F) is a binarization block diagram after (a) and (c) are overlapped and XOR-processed.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Image processing device 2 ... Image input device 3 ... Image display TV monitor 4 ... Keyboard 5 ... Mouse 10 ... Sample image or test print image 12 ... Block image divided by block size (WxH) 14 ... Block One size larger than the size (2 to 6 pixels)
Block image showing area

Claims (2)

[Claims] A picture pattern of a printed matter is fetched from a detection unit (image input unit) for each pixel, and the detected picture information of each pixel is compared with reference picture information of a corresponding pixel inputted in advance to print the picture. (1) a step of storing a pattern of a test sample serving as a reference in a reference memory as reference pattern information, and storing image information of a pattern of a test print in an image memory. (2) At least two patterns for positioning are selected from the reference pattern information to determine the position coordinates, the position coordinates of the pattern for positioning are determined from the image information of the pattern on the inspection print, and coordinate conversion is performed based on the position coordinates of both. Process. (3) A step of extracting the edge components of the picture of each of the test sample and the test print. (4) a step of converting the edge component of the picture into a binary image. (5) a step of dividing the binarized image into blocks each having a size (W pixels × H pixels). (6) A step of obtaining a position that most overlaps with the reference pattern information of the inspection sample while shifting the position of the image information of the pattern of the inspection printed matter by one pixel in each block unit. (7) A step of comparing the reference pattern information of the inspection sample and the image information of the inspection printed matter for each pixel at the position where the inspection is most overlapped, and detecting and displaying a difference between the two. And a method for inspecting printed matter. 2. An image input device for storing a pattern of an inspection sample as a reference in a reference memory as reference pattern information, and storing image information of a pattern of an inspection print in an image memory.
A coordinate conversion unit that has a reference memory and an image memory for storing the reference pattern information and the image information of the pattern of the inspection print, respectively, selects at least two points of the positioning pattern, obtains position coordinates, and performs coordinate conversion; Edge extracting means for extracting an edge component of a picture, binarized image converting means for converting the edge component of the picture into a binarized image, and converting the binarized image into blocks each having a size (W pixels × H pixels). A block dividing means for dividing, an optimal position detecting means for finding a position which overlaps the reference pattern information of the test sample by shifting the position of the image information of the pattern of the test printed matter by one pixel for each block, and an inspection at the position where the pattern overlaps most. An image processing apparatus having a difference detecting means for comparing the reference pattern information of the sample with the image information of the test print for each pixel and detecting a difference between the two; Enter the command to execute processing for displaying, and displays the differences between them, and inspection and an image display device for displaying the results, the inspection apparatus of printed matter characterized by comprising.