JP3597387B2 - Image defect detection method, image defect repair method, image defect detection device, and computer-readable recording medium - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a technique for detecting and repairing a line break defect in an image.
[0002]
[Prior art]
When a closed area surrounded by a line or the like in an image is filled (separated), if a line break occurs, a color leaks out from the line break point and the unpainted area extends to an unintended area. For example, when an original image is read from a scanner when preparing an image for a composition, a thin line may be blurred or a line may be broken due to the influence of dust. In order to prevent unintentional filling processing due to such a line break defect, conventionally, an operator has previously performed an operation of finding and repairing a line break while enlarging and displaying a closed area to be filled.
[0003]
On the other hand, research is also being conducted on a method for automatically detecting a broken line portion, such as the technique disclosed in Japanese Patent Application Laid-Open No. 4-153889. In the technique disclosed in Japanese Patent Application Laid-Open No. 4-153889, a line drawing to be inspected is subjected to thinning processing to extract an end point, and another end point is searched for in a predetermined range in the direction in which the line extends in the vicinity of the end point. Is detected, the gap between the end points is detected as a line break defect.
[0004]
Further, when a line break defect is detected, it is conceivable to perform a process of thickening the line on the original line drawing, and further perform a process of thinning the line to repair the line break defect.
[0005]
[Problems to be solved by the invention]
By the way, in the technique disclosed in Japanese Patent Application Laid-Open No. 4-153889, it is necessary to set in advance the distance between end points to be detected as line break defects. Such settings need to be changed depending on the type of image, and an AI-like technique is required to perform the setting efficiently. However, even if an AI-like setting according to the type of image is performed, detection of line break defects in all images cannot be performed. As a result, each time an operator finds a broken line defect, the operator must enlarge and repair the defective portion.
[0006]
A similar problem also occurs in the restoration process. That is, when performing a fattening process on a line in a line drawing, it is necessary to set a fattening width. However, an appropriate fattening width varies depending on the type of an image. However, it is impossible to perform a fattening process suitable for any image. Further, even if a thickening width suitable for repairing a broken line defect can be set, the thickened lines may come into contact with each other to generate a new defect. Therefore, every time an operator finds a defect, it must expand and repair it.
[0007]
SUMMARY OF THE INVENTION The present invention has been made in consideration of the above problems, and has as its object to detect only a line break defect necessary for a painting operation or the like and reduce an operator's work load.
[0008]
[Means for Solving the Problems]
The invention according to claim 1 is an image defect detection method for detecting a line break defect in a target image, wherein (a) generating a display image by thinning out the target image for display on a display unit; b) becomes a detection target of a line break defect in the target imagecolorA step of generating a reference image from the target image while preserving the connectivity of the insignificant part that is a part other than the significant part that is a part, and (c) comparing the display image and the reference image with each other. Detecting a line break defect in a significant portion in the target image.
[0009]
The invention according to claim 2 is the image defect detection method according to claim 1, wherein the step (b) includes a step of thinning out the target image while preserving the connectivity of the insignificant part.
[0010]
The invention according to claim 3 is the image defect detection method according to claim 1 or 2, wherein the step (b) is performed by using an insignificant part in the reference image.ThinA step of performing a linearization process, wherein the step (c) cuts off a position in the target image corresponding to both a significant portion in the display image and a thinning portion in the reference image. A step of obtaining the position as a defect.
[0011]
The invention according to claim 4 is the image defect detection method according to claim 1 or 2, wherein the step (a) includes a step of performing a thinning process on a significant part of the display image. (b) is a significant part of the reference imageTo dA step of performing edge extraction processing, wherein the step (c) draws a position in the target image corresponding to both a thinned portion in the display image and an edge portion in the reference image. A step of obtaining the position of the cut defect.
[0012]
According to a fifth aspect of the present invention, there is provided an image defect repairing method for repairing a line defect detected by the image defect detecting method according to any one of the first to fourth aspects, wherein the line defect detected in the target image is provided. Extracting an image in the vicinity of the existence position as a detection region image, labeling a significant portion in the detection region image, and thinning out the detection region image for display, thereby connecting the significant portion by thinning. A step of obtaining a relationship; and a step of connecting significant parts in the detection area image connected by the thinning.
[0013]
The invention according to claim 6 is an image defect detection device that detects a line break defect in a target image, wherein the unit generates a display image by thinning out the target image for display on a display unit; Of wire break defectscolorMeans for generating a reference image from the target image while preserving the connectivity of the insignificant part that is a part other than the significant part that is a part, and comparing the display image and the reference image with each other in the target image. Means for detecting a line break defect in a significant portion of
[0014]
8. The computer-readable recording medium according to claim 7, wherein said computer-readable recording medium stores a program for detecting a line break defect in a target image by a computer, said program being stored on a computer for display on a display unit. Generating a display image by thinning out, and detecting a line break defect in the target image.colorA step of generating a reference image from the target image while preserving the connectivity of the insignificant part that is a part other than the significant part that is a part, and comparing the display image and the reference image with each other in the target image The step of detecting a line break defect in a significant portion.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
<1. Principle of the Invention>
FIG. 1 is a diagram showing the principle of detecting a line break defect in the present invention. In FIG. 1, reference numeral 10 indicates a target image from which a line break defect is to be detected. In the target image 10, a portion indicated by a parallel oblique line denoted by reference numeral 11 indicates a color portion (hereinafter, referred to as a “significant portion”) to be detected as a line break defect, and the other portions denoted by reference numeral 12 are backgrounds. (Hereinafter, referred to as “insignificant part”). For example, in a line drawing in which a pattern is drawn with a black line on a white background, a black line portion becomes a significant portion, and a white background portion becomes an insignificant portion.
[0016]
Reference numeral 13 indicates a line break defect in the significant portion 11. Therefore, when the filling process of the region surrounded by the substantially rectangular significant portion 11 is performed on the target image 10 shown in FIG. Processing will be extended.
[0017]
Reference numeral 20 denotes a display image thinned out to display the target image 10 on a display unit such as a display. When the target image 10 is a fine image for printing or the like, the entire target image 10 cannot be displayed. Therefore, the display is performed after the target image 10 is thinned out for display. For example, the display image 20 is generated by replacing 2 × 2 pixels in the target image 10 with one pixel (for example, one pixel at the upper left corner). An instruction of the filling process or the like by the operator is performed based on the display image 20.
[0018]
In the display image 20 shown in FIG. 1, the portions shaded in parallel indicate significant portions 21 corresponding to the significant portions 11 in the target image 10, and the other portions correspond to the insignificant portions 12 in the target image 10. The insignificant part 22 is shown. As shown in FIG. 1, the display thinning process is a simple thinning process. Therefore, the line break defect 13 which is a small gap disappears by the thinning process and does not appear in the display image 20. That is, the operator cannot recognize the line break defect 13 from the display image 20.
[0019]
Therefore, the present invention compares the display image 20 with the target image 10 (or a later-described reference image 30 generated from the target image 10) and detects only line break defects that cannot be grasped by the operator. I have to. As a result, only the necessary line break defect is detected to reduce the work load of the operator.
[0020]
As described above, the present invention is based on comparing the target image 10 with the display image 20. However, when the target image 10 is a high-definition image for printing or the like, the comparison is rationally and appropriately performed. In order to perform this, it is preferable to prepare a reference image 30 that has been subjected to the thinning process in a state where the line break defect 13 in the target image 10 is stored.
[0021]
In FIG. 1, reference numeral 30 denotes a reference image that has been subjected to a thinning process in a state where the line break defect 13 in the target image 10 is stored. The thinning-out processing in a state where the line break defect 13 is stored is realized by performing the thinning-out processing in a state where the connectivity of the insignificant part 12 is stored. Here, “perform the thinning process with the connectivity preserved” means that the area in which the connectivity is preserved is not divided by the thinning process, and it does not matter whether the two areas are connected or not. Refers to thinning processing. In the following description, the same meaning is used.
[0022]
As a specific example of the thinning process in which the connectivity is preserved, for example, in order to maintain the connectivity of the insignificant part, if at least one pixel of the insignificant part exists in 2 × 2 pixels, the 2 × 2 pixel Is replaced with one pixel indicating an insignificant part. Thus, the thinning process can be performed without dividing the insignificant part.
[0023]
In the reference image 30 of FIG. 1, reference numeral 31 denotes a significant part corresponding to the significant part 11 in the target image 10, reference numeral 32 denotes an insignificant part corresponding to the insignificant part 12 in the target image 10, and reference numeral 33 denotes the target. A line break defect corresponding to the line break defect 13 in the image 10 is shown.
[0024]
When the display image 20 and the reference image 30 are generated from the target image 10, these images are compared, and the difference is determined as a line break defect detection position. That is, when a certain position of the display image 20 is included in the significant part 21 and a corresponding position of the reference image 30 is included in the insignificant part 32, this position is obtained as the detection position of the line break defect.
[0025]
For example, when the reference image 30 is required to be the same size as the display image 20, an exclusive OR image 40 of the display image 20 and the reference image 30 is determined, and the exclusive OR image 40 is obtained. The position of the black portion indicated as the difference between the two images in the logical sum image 40 is the defect detection position 41.
[0026]
As described above, the thinning process in which the connectivity of the insignificant part is stored is a process for easily comparing the display image 20 and the reference image 30, and the thinning rate (reduction) of the display image 20 is reduced. By referring to the (ratio), it is possible to determine to which position of the target image 10 each pixel of the display image 20 corresponds, so that the target image 10 can be used as it is as the reference image 30. .
[0027]
<2. Device configuration>
FIG. 2 is a diagram showing a configuration of an apparatus 100 for detecting and repairing a line break defect according to the present invention. As shown in FIG. 2, the apparatus 100 is configured to use a normal computer system (hereinafter, referred to as a “computer”), which executes a variety of arithmetic instructions, and stores a target image to be detected. A dedicated image data storage memory 102, a memory 103 serving as a storage area for data and programs, a reading unit 104 for reading programs and the like from a recording medium 141, an input unit 105 such as a keyboard and a mouse for receiving input from an operator, an image for an operator, etc. , A display control unit 161 which is an interface for controlling the display operation of the display unit 106, a display memory 107 for storing an image displayed on the display unit 106, and a target image in a format such as a run length. The fixed disk 108 stored as the image data 181 is appropriately Such as via a Eisu (I / F) has be connected to the bus line 100B.
[0028]
The memory 103 further includes a line break data storage memory 131 for storing a defect position as line break data, an image calculation memory 132 serving as a calculation area for an image, and a program storage memory 133 for storing a program 134 which is an operation content of the apparatus 100. Is divided into
[0029]
The program 134 is once stored in the fixed disk 108 from the recording medium 141 via the reading unit 104 in advance, and then read into the program storage memory 133 as needed. Note that the program 134 may be loaded into a computer via a separately provided communication device or the like.
[0030]
FIG. 3 is a block diagram showing the configuration of the device 100 viewed from the operation aspect. 3, the same components as those shown in FIG. 2 are denoted by the same reference numerals. Among these components, an image display unit 135 that generates display contents of an image on the display unit 106, a line break detection unit 136 that detects line breaks, and a line break instruction figure display unit 137 that displays an instruction figure at a defect detection position. , And a line break repairing unit 138 that repairs a line break defect has a configuration indicating an operation realized by the CPU 101 executing an operation command according to the program 134, and these configurations are realized by a dedicated electric circuit. Is also good.
[0031]
<3. Overall operation>
Next, an overview of the overall operation of the device 100 will be described with reference to the configuration shown in FIG.
[0032]
4 and 5 are flowcharts showing the flow of the entire operation of the apparatus 100. The apparatus 100 first reads the image data 181 in the fixed disk 108 into the image data storage memory 102 as a prerequisite operation (FIG. 4: step S11). As a result, the image data 181 is expanded in the image data storage memory 102, and a high-definition multicolor target image is prepared. The development of the image data 181 may be performed at any time as needed.
[0033]
When the target image is prepared, the image display unit 135 thins out the target image for display on the display unit 106 to generate a display multicolor image (step S12). As described above, a process of simply replacing 2 × 2 pixels with any one pixel is performed as the display thinning. Then, the generated display multicolor image is displayed on the display unit 106 via the display control unit 161 and stored in the display memory 107 (step S13).
[0034]
When the multicolor image for display is generated, the apparatus 100 receives an input of a detection area of a line break defect and a target color of the line break defect from the operator (step S14). The target color of the line break defect is, for example, when performing a filling process using a red portion as a frame for a multicolor image, if red is specified as a target color, the red portion is regarded as a significant portion that is a defect detection target. The other color portions are regarded as insignificant portions which are not the defect detection targets. The input of the target color is performed by the computer displaying the color palette used for the image on the display unit 106 and the operator selecting and instructing the target color from the color palette via the input unit 105 such as a mouse.
[0035]
Note that, in the processing described below of the apparatus 100, an image in which the target color is specified is handled as a binary image in which the pixel value of the significant part is 1 (black) and the pixel value of the insignificant part is 0 (white) in principle. .
[0036]
When an input from the operator is received, a line break defect of a significant portion which is a target color is detected (step S15). The detection of the line break defect is performed by the image calculation memory 132 and the line break detection unit 136 based on the target image in the image data storage memory 102 and the display multicolor image in the display memory 107. Is stored in the line break data storage memory 131, and details of the operation will be described later.
[0037]
If a line break defect is detected, the line break instruction graphic display unit 137 displays a figure indicating the position of the line break defect on the image displayed on the display unit 106 (steps S16 and S17). The line break instruction figure may be a figure connecting the character “line break” and the defect position by a leader line, or a simple “X” mark may be displayed at the defect position. If no line break defect is detected, the process ends.
[0038]
Next, the operator judges whether or not to repair the defect position while referring to the line break defect designating graphic displayed on the display unit 106 (FIG. 5: step S21). Another type of manual restoration is input (step S22). When the automatic repair is performed, the line break repairing unit 138 performs the automatic repair described below with reference to the contents of the line break data storage memory 131 (step S23). The line segment is drawn to restore the filled area to a closed area (step S24).
[0039]
The operation of the apparatus 100 is completed when the necessity of the restoration and the completion of the necessary restoration are confirmed for the detection position of each line break defect (step S25).
[0040]
<4. Detection operation example 1>
Next, the detection operation in step S15 in FIG. 4 will be described. FIG. 6 is a conceptual diagram showing the state of the detection operation, and FIGS. 7 and 8 are flowcharts showing the flow of the detection operation.
[0041]
In the detection operation of the line break detection unit 136, first, a binarization process in which a region of a target color of the display multicolor image is a significant portion having a pixel value of 1 and other regions are insignificant portions having a pixel value of 0 is performed. Is performed to generate a binary image for display (FIG. 7: step S31). On the other hand, the same binarization processing is performed on a high-definition target image to generate a target binary image (step S32). The display binary image may be generated by thinning out the target binary image for display (step S33).
[0042]
In FIG. 6, reference numeral 201 denotes a target binary image, and reference numeral 202 denotes a display binary image.
[0043]
Next, the target binary image 201 is inverted in black and white to generate an inverted image (step S34). In FIG. 6, reference numeral 203 denotes an inverted image. As a result, the significant portion becomes a white portion having a pixel value of 0, and the insignificant portion becomes a black portion having a pixel value of 1.
[0044]
When the inverted image 203 is generated, the insignificant part (black part) of the inverted image 203 is subjected to thinning processing (step S35). In FIG. 6, reference numeral 204 illustrates an image after thinning. Then, the thinning process is performed while preserving the connectivity of the thinned portion which is the black portion in the image 204, and the reference image 205 having the same size as the display binary image 202 is generated (step S36).
[0045]
When the display binary image 202 and the reference image 205 are generated, a logical sum image 206 of these images is obtained (FIG. 8: step S41), and a set of pixels whose pixel values in the logical sum image 206 are 1 The area is set as a defect detection position (step S42). By multiplying the defect position detected in this way by the thinning rate, the coordinates of the line-break defect in the target image are obtained, and the line-break data storage memory 131 stores the coordinates of the line-break defect in the repair work described later. The coordinates are saved.
[0046]
Note that, even when the reference image 205 is not generated in the same size as the display binary image 202, the logical sum image 206 can be generated by calculating the logical sum at the position corresponding to both images. . Further, the logical sum image 206 may be generated by comparing the binary image for display 202 and the image 204 after thinning with reference to the thinning rate by omitting step S36.
[0047]
In the defect detection operation shown in FIG. 6, thinning processing is performed on the insignificant part in the target binary image 201, and thinning processing is performed while preserving the connectivity of the thinned insignificant part. . That is, these processes correspond to the generation process of the reference image 30 shown in FIG. 1, and the generation of the OR image after the inversion process corresponds to the generation of the exclusive OR image 40 in FIG. The purpose of performing the thinning processing is to remove the influence of the deformation of the significant part due to the difference in algorithm between the display thinning processing and the insignificant part connectivity preserving thinning processing.
[0048]
Further, in the operation shown in FIG. 6, the thinning processing in which the connectivity is preserved after the thinning is performed (steps S35 and S36), but the thinning processing of the insignificant part is performed after the thinning processing in which the connectivity of the insignificant part is preserved. May be performed.
[0049]
Further, it is not necessary to perform the thinning in step S35 until a skeleton image having a width of one pixel is obtained.
[0050]
With the above-described defect detection operation, a line break defect that disappears in the display image can be accurately detected. Accordingly, the operator can perform the defect repair work while grasping the entire target image via the display unit 106 without considering a line break defect that does not appear on the display unit 106. As a result, it is not necessary to perform an enlarging operation for searching for a defect, and it is possible to significantly reduce an operator's work load.
[0051]
<5. Detection operation example 2>
Next, another example of the detection operation in step S15 in FIG. 4 will be described. FIG. 9 is a conceptual diagram showing a state of the detection operation, and FIG. 10 and the above-mentioned FIG. 8 are flowcharts showing the flow of the detection operation.
[0052]
In the detection operation illustrated in FIG. 9, first, the binarization is performed by setting the pixel value of the significant part of the multicolor image for display to 1 and the pixel value of the insignificant part to 0 to generate a binary image for display. 10: Step S51) Also in the target image, binarization is performed by setting the pixel value of the significant part to 1 and the pixel value of the insignificant part to 0 to generate a target binary image (Step S52). The display binary image may be generated by thinning out the target binary image for display (step S53). In FIG. 9, reference numerals 301 and 302 indicate a target binary image and a display binary image, respectively.
[0053]
Next, a thinning process is performed on a significant portion (black portion) of the display binary image 302 to generate an image 303 (step S54). In the example shown in FIG. 9, since the line break defect has disappeared in the display binary image 302, the thinned image 303 has a thinned portion (black portion) that is a rectangular line.
[0054]
On the other hand, the target binary image 301 is subjected to 8-neighbor one-pixel expansion processing in which the pixel values of the pixels existing in the eight neighborhoods of each pixel of the significant part are set to 1, and an image 304 is generated (step S55). ), An image of the exclusive OR of the original target binary image 301 and the image 304 subjected to the dilation processing is obtained to generate the edge image 305 of the significant part (step S56). The reference image 306 having the same size as the display image is generated by performing the thinning process while preserving the connectivity of the edge portion (black portion) in the edge image 305 (step S57).
[0055]
Thereafter, a logical sum image 307 of the image 303 obtained by performing the thinning processing on the display binary image 302 and the reference image 306 is obtained in substantially the same manner as the processing shown in FIG. 8 (step S41). , That is, the positions corresponding to the thinned portion in the image 303 and the edge portion in the reference image 306 are obtained as the position of the line break defect (step S42).
[0056]
By multiplying the defect position detected in this way by the thinning rate, the coordinates of the line-break defect in the target image are obtained, and the line-break data storage memory 131 stores the coordinates of the line-break defect in the repair work described later. The coordinates are saved.
[0057]
In the above operation, in the image 303 after the thinning, the connectivity of the significant part due to disappearance of the line break defect when generating the display binary image 302 from the target image is preserved. Further, since the edge image 305 indicates the connectivity of the insignificant part in the target binary image 301, the generation of the reference image 306 corresponds to the thinning processing in which the connectivity of the insignificant part in FIG. . Therefore, by obtaining the logical sum image 307 of the image 303 and the reference image 306, the same processing as the comparison between the display image 20 and the reference image 30 shown in FIG. Is detected.
[0058]
Here, the purpose of obtaining the edge image 305 is to simplify the connectivity preserving thinning processing of the edge part (corresponding to the insignificant part), and the purpose of obtaining the thinned image 303 is the thinning processing. The purpose of the present invention is to eliminate the influence of the slight difference between the shape of the contour in the binary image for display 302 and the shape of the edge portion in the image for reference 306 due to the difference in the algorithm.
[0059]
Note that, similarly to the example shown in FIG. 6, the reference image 306 does not necessarily need to be thinned out to the same size as the display image 302 or the image 303 after thinning. The logical sum may be obtained at both corresponding positions to obtain the logical sum image 307. Further, step S57 may be omitted, and the logical sum image 307 may be directly obtained by referring to the thinning rate based on the image 303 obtained by thinning the display binary image and the edge image 305.
[0060]
In addition, the order of step S56 and step S57 may be reversed, and the edge extraction process is performed after the thinning process is performed while the connectivity of the insignificant part of the target binary image 301 is preserved, and the final reference Image 306 may be generated.
[0061]
Further, the method of edge extraction in step S56 is not limited to the above-described method, and various methods may be used, and the thinning processing in step S54 also needs to perform thinning to a skeleton image having a width of one pixel. There is no.
[0062]
Also in the defect detection operation described above, it is possible to accurately detect a line break defect that disappears in a display image. This allows the operator to perform a later-described defect repair process while capturing the entire target image via the display unit 106 without considering line break defects that do not appear on the display unit 106. As a result, it is not necessary to perform an enlarging operation for searching for a defect, and it is possible to significantly reduce an operator's work load.
[0063]
<6. Automatic repair operation>
Next, the automatic repair operation of the line break defect in step S23 in FIG. 5 will be described. FIG. 11 is a diagram showing a state of the automatic restoration operation, and FIG. 12 is a flowchart showing a flow of the automatic restoration operation.
[0064]
In the automatic repair operation, first, the coordinates of the position of the line break defect are read from the line break data storage memory 131, and an image in a predetermined range near the line break defect is extracted from the target binary image as a detection area image (FIG. 12: Step S61). In FIG. 11, reference numerals 401 and 402 indicate the original target binary image and the detection area image, respectively. Since the width of the line break defect is approximately determined from the ratio of the size of the target image to the size of the display image, the detection region image is extracted within a range of about 2 to 5 times the width of the line break defect.
[0065]
Further, labeling is performed on each region of the significant portion (black portion) in the detection region image 402 (step S62). It is assumed that the significant part areas 402a, 402b, and 402c shown in FIG. Note that the labeling is performed on each of the significant portion regions that exist independently in the detection region image 402, and it does not matter how the significant portion regions are connected outside the detection region image 402.
[0066]
Next, thinning similar to thinning for display is performed on the detection area image 402 (step S63). Since only line defects that disappear by the display thinning process are detected, the line defect in the detection area image 402 disappears by the display thinning. Therefore, in the thinned image 403, at least two of the significant part regions 402a, 402b, and 402c are connected. In the image 403 after thinning shown in FIG. 11, an example is shown in which significant part regions 402a, 402b, and 402c are connected to each other. Thus, it is determined how the line break defect that does not appear in the display unit 106 exists, that is, how to link the significant part regions, and the connection relation is stored in the connection table 404. (Step S64).
[0067]
Note that the connection relationship between the significant regions by the display thinning is obtained by scanning a 2 × 2 pixel window on the image 403 after the thinning. That is, when there is a state in which pixels corresponding to different significant part areas are simultaneously included in the window, it is determined that these significant part areas are connected by thinning.
[0068]
The connection table 404 shown in FIG. 11 shows that the significant area 402a labeled 1 is to be connected to the significant areas 402b and 402c labeled 2 and 3, respectively. Note that in the connection table 404 shown in FIG. 11, the storage of the connection relationship between the significant part regions 402b and 402c to which the labels 2 and 3 are attached is directly or indirectly overlapped with the connection relationship of the label 1 so that the connection is performed. Relationships are not remembered.
[0069]
When the connection relation of each significant part area is obtained, the edge of the significant part of the detection area image 402 is extracted to generate the detection area edge image 405 (step S65), and refer to the connection table 404 and the detection area edge image 405. While drawing, a line segment is drawn at a position that is the shortest distance between the significant part areas to be connected (steps S66 and S67).
[0070]
Here, the detection area edge image 405 is used to simply obtain the drawing position of a line segment connecting the significant part areas to be connected with the shortest distance. Also, in FIG. 11, reference numeral 406 denotes the restored detection area image 406, and reference numerals 461 and 462 denote line segments drawn between the significant part areas 402a and 402b and between the significant part areas 402a and 402c, respectively. Shows the line segment drawn in the.
[0071]
With the automatic repair operation as described above, a line break defect that does not appear on the display unit 106 can be properly repaired. Accordingly, the operator can repair the line break defect while referring to the display multicolor image thinned out for display displayed on the display unit 106, that is, while grasping the entire target image. It is not always necessary to perform the operation of enlarging and displaying the defect position (however, the detection area image may be enlarged and displayed for the operator's confirmation). As a result, the work load on the operator can be greatly reduced, and the process can be smoothly shifted to the process of filling (separating) the target image after the repair of the line break defect.
[0072]
The apparatus 100 for detecting and repairing a line break defect according to the present invention has been described above. However, the apparatus 100 detects and repairs only a line break defect that is difficult for an operator to grasp by thinning processing when displaying a target image. It has become. That is, the "line break defect" is treated as being positioned as a line break defect that does not appear on the display. As a result, it is possible to reduce the work load of the operator for the enlargement processing and the like, and to shorten the work time.
[0073]
Further, since only line break defects that cannot be grasped on the display are detected, it is not necessary to set a numerical value for detecting a line break defect as in the related art, and complicated setting and AI determination according to the type of image can be performed. It becomes unnecessary. Further, since the detection operation is not based on the shape of the line break defect such as the end point of the line drawing, the line break defect can be detected without depending on the shape of the line break defect.
[0074]
In addition, since the apparatus 100 performs processing on a multicolor target image by distinguishing a significant part to be subjected to line break defect detection from other insignificant parts, the apparatus 100 flexibly supports a multicolor image. Have been able to be.
[0075]
<7. Modification>
Although the apparatus 100 according to the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible.
[0076]
For example, in the above embodiment, a multicolor high-definition image is used as the target image, but the target image may be a binary image.
[0077]
Further, the thinning process for display may be any thinning process. For example, the pixel value may be thinned out to one pixel having the average value of the pixel values of 4 × 4 pixels as the pixel value.
[0078]
Further, the detection operation in the above embodiment is a specific example, and various modifications can be made in accordance with the principle of the invention. For example, in the example shown in FIG. 6, it is also possible to omit the thinning processing of the insignificant part and the thinning processing in which the connectivity of the thinning part is stored. In the example shown in FIG. 9 as well, the edge extraction processing and thinning of the significant part can be omitted as necessary.
[0079]
Further, in the example of the restoration operation shown in FIG. 11, the edge extraction processing of the significant part area can be omitted.
[0080]
Further, in order to reduce the time required for the thinning processing, the thinning processing in each step may be performed for several pixels.
[0081]
【The invention's effect】
According to the first to seventh aspects of the present invention, a line break defect that cannot be grasped from the display image can be detected, so that only a line break defect necessary for the operator can be detected. Thereby, the work load of the operator is reduced.
[0082]
According to the second aspect of the present invention, since the thinning is performed when the reference image is generated, the display image and the reference image can be easily compared.
[0083]
Further, according to the third and fourth aspects of the present invention, line break defects can be accurately detected.
[0084]
Further, according to the fifth aspect of the present invention, it is possible to appropriately repair a line break defect that cannot be grasped from a display image.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram showing the principle of the invention.
FIG. 2 is a block diagram showing a configuration of an apparatus according to the present invention.
FIG. 3 is a block diagram showing the configuration of the device as viewed from the operation.
FIG. 4 is a flowchart showing a flow of an overall operation.
FIG. 5 is a flowchart showing a flow of an overall operation.
FIG. 6 is a conceptual diagram showing a state of a detection operation.
FIG. 7 is a flowchart showing a flow of a detection operation.
FIG. 8 is a flowchart showing a flow of a detection operation.
FIG. 9 is a conceptual diagram showing a state of another example of the detection operation.
FIG. 10 is a flowchart showing an operation flow in another example of the detection operation.
FIG. 11 is a conceptual diagram showing a state of a repair operation.
FIG. 12 is a flowchart showing a flow of a repair operation.
[Explanation of symbols]
10, 201, 301, 401 Target binary image
11 significant part
12 Insignificant part
13 Line break defect
20, 202, 302 Display image
30, 205, 306 Reference image
100 devices
101 CPU
103 memory
106 Display
108 fixed disk
132 Image calculation memory
134 programs
135 Image display section
136 Line break detector
141 Recording Medium
402 Detection area image
S12, S31 to S36, S41, S42, S51 to S57, S61 to S67 steps

Claims (7)

An image defect detection method for detecting a line break defect in a target image,
(a) generating a display image by thinning the target image for display on a display unit,
(b) generating a reference image from the target image while preserving the connectivity of the insignificant part that is a part other than the significant part that is a part of the color to be detected for a line break defect in the target image,
(c) comparing the display image and the reference image to detect a line break defect in a significant portion in the target image,
An image defect detection method, comprising: The image defect detection method according to claim 1,
The step (b),
Thinning the target image while preserving the connectivity of the insignificant part,
An image defect detection method, comprising: The image defect detection method according to claim 1 or 2,
The step (b),
Performing thinning processing on the insignificant part in the reference image,
Has,
The step (c) comprises:
A step of obtaining a position in the target image corresponding to both the significant portion in the display image and the thinned portion in the reference image as a position where a line break defect exists,
An image defect detection method, comprising: The image defect detection method according to claim 1 or 2,
The step (a),
Performing a thinning process on a significant portion of the display image,
Has,
The step (b),
Performing an edge extraction process on a significant portion of the reference image,
Has,
The step (c) comprises:
A step of obtaining a position in the target image corresponding to both the thinned portion in the display image and the edge portion in the reference image as a position where a line break defect exists,
An image defect detection method, comprising: An image defect repairing method for repairing a line break defect detected by the image defect detecting method according to claim 1,
Extracting an image near the position of the line break defect detected in the target image as a detection region image,
Labeling a significant part in the detection area image;
By thinning the detection area image for display, a step of determining the connection relationship of the significant part by thinning,
Connecting between significant parts in the detection area image connected by the thinning,
A method for repairing an image defect, comprising: An image defect detection device that detects a line break defect in a target image,
Means for generating a display image by thinning the target image for display on a display unit,
Means for generating a reference image from the target image while preserving the connectivity of the insignificant part that is a part other than the significant part that is the color part to be detected for the line break defect in the target image,
Means for comparing the display image and the reference image to detect a line break defect in a significant portion in the target image,
An image defect detection device comprising: A computer-readable recording medium recording a program for detecting a line break defect in a target image by a computer, wherein the program stores
Generating a display image by thinning out the target image for display on a display unit;
A step of generating a reference image from the target image while preserving the connectivity of an insignificant part that is a part other than a significant part that is a part of a color to be detected for a line break defect in the target image,
A step of comparing the display image and the reference image to detect a line break defect in a significant portion in the target image,
And a computer-readable recording medium for executing at least the following.

Images