JP7082426B2 - Image processing device and image processing method - Google Patents

Description

The present invention relates to an image processing apparatus and an image processing method, and particularly to image processing of a screen printing plate.

In a printing plate for screen printing (screen printing plate), a part of an emulsion film formed by applying a mesh thread made of metal or nylon on a screen mesh having a mesh structure (mesh structure) is an object to be printed. It has a configuration in which an opening (screen opening) corresponds to a pattern (printing pattern) to be formed.

For example, when a mask pattern made of a resist material is to be formed by screen printing on a copper-pasted plate in order to provide a wiring pattern such as a wiring (linear pattern) or a land (circular pattern) on a printed circuit board, those wiring patterns are used. A screen printing plate provided with a screen opening corresponding to the above is prepared. Then, after forming a mask pattern made of a resist material on a copper-pasted plate by screen printing using the screen printing plate, unnecessary copper is removed by an etching process, and further, the mask pattern is removed to obtain a desired pattern. A printed circuit board on which a wiring pattern is formed can be obtained.

When a printed circuit board is manufactured by such a procedure, it is required that the line width and the diameter of the circle in this wiring pattern are accurate in terms of process control. Therefore, when producing a screen printing plate, the screen opening needs to be provided with high accuracy.

On the other hand, when the screen printing plate is used repeatedly, the shape of the screen opening may be deformed. Therefore, from the viewpoint of ensuring the dimensional accuracy of the wiring pattern, the dimensions of the screen opening provided in the screen printing plate used to form the wiring pattern are important measurement control items (inspection items) in the manufacture of the printed circuit board. Therefore, a highly accurate measurement method (inspection method) is required.

From the viewpoint of excluding human measurement errors, it is desirable to carry out automatic measurement by image processing. However, when the mask pattern provided on the screen mesh is to be automatically measured by image processing, the image of the screen mesh existing in the pattern opening may affect as noise.

A method of removing a screen mesh portion from an original image by applying an expansion / contraction treatment (opening treatment and closing treatment) to a monochrome acquired image is already known (see, for example, Patent Document 1).

Japanese Patent No. 4652717

The technique disclosed in Patent Document 1 is one of the general image processing methods for noise removal in principle, but it can erase a pattern close to the size of the mesh line width as noise and the opening. There is a problem that the original shape of the opening is deformed by the processing or the closing process.

Due to the difficulty of such automatic measurement, manual measurement by visual inspection is still dominant in the measurement (inspection) of the pattern shape of the screen printing plate. Manual measurement has a problem that the inspection efficiency is low and it is difficult to secure the accuracy because individual differences may occur depending on the measurer.

The present invention has been made in view of the above problems, and is an image processing method suitable for an image used for measuring (inspecting) the pattern shape (opening shape) of a screen printing plate, and an image processing for realizing the method. The purpose is to provide the device.

In order to solve the above problem, the invention of claim 1 is an image processing apparatus for extracting a predetermined region matching the existence position of the line pattern from a processing target image including an image of the line pattern, and the processing target image. Is displayed, an operation unit that can perform a predetermined operation on the image to be processed, an extraction standard setting processing unit that sets an extraction standard when extracting the predetermined area, and an extraction standard based on the extraction standard. A region extraction processing unit for extracting the predetermined region from the image to be processed is provided, and the predetermined region is a region in which the hue is within a predetermined hue value range and is a base of the line pattern. It is a region having irregular shading due to processing marks existing in the portion, and the extraction standard setting processing unit operates the operation unit while visually recognizing the processing target image displayed on the display unit. By doing so, a range selection unit that enables selection of an arbitrary pixel range from the image to be processed, and a hue value of pixels belonging to the arbitrary pixel range selected by the operation of the operation unit by the operator. Has a hue range specifying unit for specifying the range of the above based on the data of the image to be processed described in the HSV color coordinate system, and is performed by the operator as the selection of the arbitrary pixel range. In response to the selection of a small area including the shade from the area visually recognized by the operator when the image to be processed is included in the predetermined area, the image is included in the small area specified by the hue range specifying unit. The range obtained by removing the entire range of the hue value range of the pixels or the range obtained by removing the predetermined ratios of the upper and lower limits set in advance is set as the extraction reference, and the area extraction processing unit sets the hue in the image to be processed. It has a pixel specifying unit for specifying a pixel whose value corresponds to the extraction standard, and is characterized in that a region consisting of all the pixels specified by the pixel specifying unit is extracted as the predetermined region. ..

The invention of claim 2 is the image processing apparatus according to claim 1, which is an imaging processing unit that captures a predetermined imaging target and generates imaging data described in an RGB color coordinate system, and an RGB color coordinate system. Further provided with a color space conversion unit capable of converting the data of the processing target image described in the above into the data of the HSV color coordinate system, the data of the processing target image described in the HSV color coordinate system is a predetermined imaging. It is characterized in that it is obtained by converting the image pickup data obtained by taking an image of an object by the image pickup processing unit by the color space conversion unit.

The invention of claim 3 is the image processing apparatus according to claim 1 or 2 , wherein the region extraction processing unit applies a hue value of a pixel specified by the pixel specifying unit to the image to be processed. It is characterized by further having a hue replacement processing unit that replaces a hue value different from the hue value that gives an image of the predetermined region.

The invention of claim 4 is the image processing apparatus according to any one of claims 1 to 3 , wherein the whole or end of the predetermined area extracted by the area extraction processing unit is the display unit. It is characterized in that it is highlighted in the processed image displayed on the screen.

The invention according to claim 5 is an image processing method for extracting a predetermined region matching the existence position of the line pattern from a processing target image obtained by imaging a predetermined imaging target including an image of the line pattern. An extraction standard setting step for setting an extraction standard when extracting a predetermined region and a region extraction step for extracting the predetermined region from the processing target image based on the extraction standard are provided, and the predetermined imaging target is provided. However, a predetermined pattern is formed on the base surface, and the predetermined region is a region in which the hue is included in the image of the predetermined pattern in the image to be processed and the hue is within the range of the predetermined hue value. It is a region having irregular shading due to processing marks existing in the base portion of the line pattern , and the extraction standard setting step is the processing target displayed on a predetermined display means by the operator. By operating a predetermined operating means while visually recognizing an image, a range selection for selecting an arbitrary small area including the shade from the area visually recognized by the operator when the image to be processed is included in the predetermined area. It has a step and a hue range specifying step of specifying a range of hue values of pixels belonging to the arbitrary small area based on the data of the processed image described in the HSV color coordinate system. The range obtained by removing the entire range of the hue values of the pixels belonging to the arbitrary small area or the predetermined ratios of the upper and lower limits set in advance specified in the hue range specifying step is set as the extraction reference. The region extraction step includes a pixel specifying step of specifying a pixel whose hue value corresponds to the extraction reference in the image to be processed, and a region consisting of all the pixels specified in the pixel specifying step. , It is characterized in that it is extracted as the predetermined region.

The invention of claim 6 is the image processing method according to claim 5 , which is an imaging processing step of imaging a predetermined imaging target and generating imaging data described in an RGB color coordinate system, and an RGB color coordinate. Further comprising a color space conversion step capable of converting the data of the processing target image described in the system into the data of the HSV color coordinate system, the data of the processing target image described in the HSV color coordinate system is a predetermined. It is characterized in that the image pickup data obtained by imaging the image pickup object in the image pickup processing step is obtained by being converted in the color space conversion step.

According to the first to sixth aspects of the invention, a simple process of selecting a part of a region where the hue is within a predetermined hue value range and the shade is visible is suitable. Can be extracted. By performing the processing based on the hue, the influence of the difference in brightness and vividness in the portion where the shade is repeated, which is a problem when the processing based on the color density value of each RGB color is performed, is suitably eliminated.

In particular, according to the inventions of claims 2 and 6 , even an image represented by data described in the RGB color coordinate system generated by imaging by a camera is described in the HSV color coordinate system. By converting to data, the extraction process can be preferably performed.

In particular, according to the invention of claim 3 , since the hue value of the extracted region is replaced with a certain hue value, it becomes easy to specify the region in the measurement process for the region. As a result, the measurement process can be performed with high accuracy.

In particular, according to the invention of claim 4 , it becomes easy to grasp the extracted region in the image to be processed.

It is a figure which shows the schematic structure of the measuring apparatus 100. It is a functional block diagram which shows the functional component provided in the measuring apparatus 100. It is a figure which shows the captured image (partially enlarged image) IM1 about an example of a screen printing plate 10. It is a figure which shows the procedure of image processing. It is a figure which shows the captured image IM2 when the rectangular area RE is selected by the drag operation of the operator with the mouse provided as the operation part 133. It is a figure which shows the histogram HG1 about the hue value of all the pixels belonging to the rectangular region RE shown in FIG. It is a figure which illustrates the binary image IM3a which shows the range occupied by the pixel specified by the corresponding pixel identification part 191 in black, and the image IM3b after replacement by a hue replacement processing part 192. It is a figure which shows the histogram HG2 about all the pixels which make up the captured image IM1 before replacement. It is a figure which shows the image IM1r obtained by extracting only the R component from the RGB image pickup data D0 which gives the image pickup image IM1, and further binarizing it. It is a figure which shows the state when the same image processing as the embodiment is applied to the line pattern 51 formed on the metal plate 50.

<Outline of the invention>
FIG. 1 is a diagram showing a schematic configuration of a measuring device 100 that functions as an image processing device capable of performing image processing according to the present embodiment. FIG. 2 is a functional block diagram showing functional components included in the measuring device 100. In the present embodiment, the image processing for the screen printing plate 10 to be processed and the configuration of the measuring device 100 related thereto, which are performed in the measuring device 100, will be described.

The measuring device 100 includes a stage 101 on which the screen printing plate 10 to be measured is placed, a camera body 111, and an imaging unit 110 (FIG. 2) that captures the screen printing plate 10 mounted on the stage 101. A lighting unit 120 (FIG. 2) having three types of lighting means (coaxial lighting 121, ring lighting 122, and transmitted lighting 123) and capable of irradiating the screen printing plate 10 with illumination light in various modes, and measurement. It mainly includes a processing device 130 that controls the operation of each part of the device 100 and performs arithmetic processing necessary for specifying the mesh position. The details of the configuration of the measuring device 100 will be described later.

FIG. 3 is a diagram showing a captured image (partially enlarged image) IM1 of an example of the screen printing plate 10. The screen printing plate 10 roughly corresponds to a pattern (printing pattern) to be formed on an object to be printed on a part of an emulsion film 2 formed in a plate frame 1 having a rectangular shape in a plan view made of a metal such as aluminum. An opening (pattern opening) 3 is provided (FIG. 1).

However, in FIG. 1, the illustration of the screen printing plate 10 is simplified. In the actual screen printing plate 10, the screen mesh 4 is stretched on the plate frame 1 as shown in FIG. The screen mesh 4 has a mesh structure (mesh structure) woven so that a large number of weft threads 4x and a large number of warp threads 4y, each of which is made of a metal or nylon wire and is separated at a predetermined interval, intersect (orthogonally) with each other. The emulsion film 2 is formed by coating on the screen mesh 4 (so to speak, using the screen mesh 4 as a base surface). Therefore, the screen mesh 4 is exposed in the opening 3. The emulsion film 2 is provided with a predetermined color. This is for ensuring the distinction from the opening 3 where the screen mesh 4 is exposed, that is, for ensuring the visibility. Although it cannot be seen from the drawings due to the limitation of the illustration, the emulsion film 2 of the screen printing plate 10 illustrated in FIG. 3 is actually visually recognized as reddish purple.

In the present embodiment, the captured image of the emulsion film 2 including the opening 3 such as the captured image IM1 illustrated in FIG. 3 is the target of image processing in the measuring device 100. The image processing according to the present embodiment is roughly intended to extract a region other than the pattern opening 3, that is, a region where the emulsion film 2 is formed, from the image of the screen printing plate 10. In other words, it is intended to specify (the pixel range that gives the image) all the pixels that give the image of the emulsion film 2 in the image pickup data of the screen printing plate 10.

The emulsion film 2 is coated and formed on the screen mesh 4, but in many cases, the screen mesh 4 is transparent to the emulsion film 2. Therefore, when the emulsion film 2 which should originally exhibit the same color tone is visually or microscopically observed, the shade (difference in color tone) due to the permeation is the warp and weft 4x and the warp and weft 4y. It is confirmed that it is repeated in each extension direction. Although the warp and weft 4x and the warp and weft 4y are arranged at approximately equal intervals, the position shift is not small, so that the shading of the emulsion film 2 is substantially periodic, but is not necessarily repeated at a constant cycle. Not at all. Further, even if the end portion of the emulsion film 2 forming the opening 3 is macroscopically straight, microscopically, it has a fine uneven shape due to the presence of the screen mesh 4. The image processing according to the present embodiment, which will be described later, is a process intended to extract a formed region of the emulsion film 2 while taking into consideration the characteristics peculiar to the emulsion film 2.

Strictly speaking, FIG. 3 is merely an image IM1 for an exemplary aspect of the screen printing plate 10, and therefore, only the emulsion film 2, the pattern opening 3, and the pattern opening 3 are present in the figure. Although it is an image of the screen mesh 4, in the present embodiment, these images may be simply referred to as an emulsion film 2, a pattern opening 3, and a screen mesh 4 for the sake of simplicity. The same shall apply in the following figures. Further, the pattern opening 3 is also referred to as a measurement target pattern.

Further, FIG. 3 shows the captured image IM1 of the screen printing plate 10 in which a large number of linear pattern openings 3 having a predetermined width are formed substantially in parallel. The arrangement is not limited to this, and patterns of various shapes such as a rectangle, a circle, and the like may be formed in various arrangements.

<Configuration of measuring device>
Next, the configuration of the measuring device 100 will be described with reference to FIGS. 1 and 2.

As shown in FIG. 1, the stage 101 is provided on the upper surface thereof so that an object to be measured (screen printing plate 10 in the case of FIG. 1, the same applies hereinafter) can be placed and fixed in a horizontal posture. For fixing, various methods such as adsorption, sticking, and teaching by a holding means (not shown) may be applied. It should be noted that, in the state where the mounting is fixed, at least the image pickup target range is, for example, glass so that the illumination light from the transmitted illumination 123 arranged below is irradiated to at least the image pickup target range by the image pickup unit 110. It is composed of transparent materials such as.

The image pickup unit 110 mainly includes a camera body 111 provided above the stage 101, and a lens barrel 112 connected to the camera body 111 and extending vertically downward. Inside the lens barrel 112, an image pickup lens and other optical systems 113 (not shown) are provided.

The camera body 111 is used with an appropriate imaging lens attached, and can capture a color image represented by an RGB color coordinate system for a subject placed on the stage 101.

As described above, the illumination unit 120 includes three types of illumination means: a coaxial illumination 121, a ring illumination 122, and a transmission illumination 123. These lighting means have different characteristics.

The coaxial illumination 121 is an illumination means that irradiates the illumination light toward the measurement object mounted on the stage 101 in such a manner that the central axis of the illumination light coincides with the central axis of the angle of view of the camera body 111. Specifically, a half mirror (not shown) is provided in the middle of the imaging optical path from the camera body 111 in the lens barrel 112 to the stage 101, and is emitted from the coaxial illumination 121 provided on the side of the lens barrel 112. The illumination light is reflected by the half mirror to be coaxial with the imaging optical path, and then is irradiated toward the object to be measured.

The ring illumination 122 is an illumination means arranged in a ring shape (annular shape) on the outer circumference of the lens barrel 112. The ring illumination 122 is arranged so as to irradiate the illumination light toward the measurement object placed on the stage 101.

The transmitted illumination 123 is an illumination means arranged below the stage 101, and irradiates the illumination light vertically upward. That is, the transmitted illumination 123 irradiates the measurement object mounted on the stage 101 with illumination light from the side opposite to the coaxial illumination 121 and the ring illumination 122. The illumination light is applied to the back surface of the screen printing plate 10 through the transparent portion of the stage 101.

The coaxial illumination 121, the ring illumination 122, and the transmission illumination 123 are appropriately selected or combined and used according to the usage mode of the measuring device 100. Preferably, the irradiation mode of the illumination light from each illumination means is determined so that at least the entire imaging range by the imaging unit 110 is uniformly illuminated.

The measuring device 100 also includes a predetermined moving mechanism 100M (FIG. 2) for moving the image pickup unit 110 and the illumination unit 120 with respect to the stage 101. By moving the image pickup unit 110 together with the illumination unit 120 by the moving mechanism 100M, it becomes possible to image an object to be measured mounted and fixed on the stage 101 at an arbitrary image pickup position in the horizontal plane and in the vertical direction. There is.

The processing device 130 includes a processing control unit 131 including a CPU 131a, a ROM 131b, and a RAM 131c, a display unit 132 including, for example, a display, an operation unit 133 including a mouse, a keyboard, a touch panel, and a storage unit including a hard disk. It is equipped with 134. The processing device 130 can be realized by, for example, a general-purpose computer.

In the measuring device 100, when a predetermined processing program (not shown) stored in the storage unit 134 is executed by the processing control unit 131, various functional components shown in FIG. 2 are virtualized in the processing control unit 131. It has come to be realized as a typical component.

Specifically, the image pickup processing unit 140, the lighting control unit 150, the movement control unit 160, the color space conversion unit 170, the extraction standard setting processing unit 180, the area extraction processing unit 190, and the measurement processing unit 200. However, it is mainly realized.

The image pickup processing unit 140 controls the image pickup by the image pickup unit 110, acquires the data (imaging data) obtained as a result of the image pickup from the camera body 111, and performs a predetermined process in the process control unit 131.

Specifically, the image pickup processing unit 140 causes the image pickup unit 110 to perform imaging of the screen printing plate 10, and describes the image pickup result by describing the color density value of each pixel based on the RGB color coordinate system (RGB color space). It is generated as the RGB image pickup data D0. Alternatively, the RGB image pickup data D0 may be generated in the image pickup unit 110 (more specifically, the camera body 111) and passed to the image pickup processing unit 140.

The image pickup processing unit 140 also plays a role of displaying the image (live view) imaged by the image pickup unit 110 on the display unit 132.

The illumination control unit 150 controls the irradiation of the illumination light by the illumination unit 120 at the time of imaging by the image pickup unit 110. The specific irradiation mode may be appropriately determined according to the shape and color of the measurement target, the position of the measurement target, and the like. For example, it is assumed that the illumination light is irradiated with a brightness such that the color density value of each pixel in the RGB image pickup data D0 is not saturated.

The movement control unit 160 controls the operation of the movement mechanism 100M that moves the image pickup unit 110 and the illumination unit 120. The control mode of the movement mechanism 100M by the movement control unit 160 includes manual control according to a movement instruction given to the movement control unit 160 by an operator (operator) of the measuring device 100 through the operation unit 133, and a preset processing program. According to this, there is an automatic control in which imaging by the imaging unit 110 and movement by the moving mechanism 100M are executed. The latter is assumed to be, for example, a case where imaging and further measurement are sequentially performed at a plurality of predetermined different locations of one measurement target.

The color space conversion unit 170 is responsible for converting a color density value described based on the RGB color coordinate system into a description based on the HSV color coordinate system (HSV color space).

The color density value of a pixel i in the RGB color coordinate system is (R, G, B) = (Ri, Gi, Bi) (0≤Ri, Gi, Bi≤255), and the maximum value of the values Ri, Gi, Bi. Is Max and Min is the minimum value, and the color density value (H, S, V) = (Hi, Si, Vi) in the HSV color coordinate system is expressed by the following known relational expression.

Hi = 60 ・ (Gi-Ri) / (Max-Min) +60 (when Min = Bi);
Hi = 60 · (Bi-Gi) / (Max-Min) +180 (when Min = Ri);
Hi = 60 · (Ri-Bi) / (Max-Min) +300 (when Min = Gi);
Hi = indefinite (when Max = Min);
Si ＝ Max－Min ；
Vi = Max.

However, for Hi obtained by the above relational expression, when Hi <0, 360 is further added, and when Hi> 360, 360 is further subtracted, and the value is regarded as Hi.

The color space conversion unit 170 converts the color density value of each pixel described in the RGB image pickup data D0 based on the RGB color coordinate system based on the above relational expression, so that the pixel values of all the pixels are HSV colors. The HSV imaging data D1 described based on the coordinate system is generated. The HSV imaging data D1 is data associated with the RGB imaging data D0 using the pixel position information (coordinates) as a key.

The value Hi of the hue H obtained by the above relational expression is a value in the range of 0 ≦ Hi ≦ 360 corresponding to the color wheel, but the standardized hue value Hn is standardized to 256 gradations. May be used instead of Hi. In this case, approximately
Hn = (256/360) ・ Hi
Is.

In the measuring device 100 according to the present embodiment, the RGB image pickup data D0 obtained by the image pickup by the image pickup unit 110 is not the same, but the HSV image pickup data D1 obtained by converting the data by the color space conversion unit 170 is the image pickup image IM1. It is used in the actual processing for extracting the formed region of the emulsion film 2 from the above. This is in consideration of the fact that the captured image IM1 has a shade (difference in color) due to the presence or absence of transmission of the screen mesh 4 as described above. Even if the captured image IM1 is obtained in a state where the illumination light is evenly irradiated as described above, there is a possibility that shading may occur due to a difference in the imaging location or location.

In the RGB image pickup data D0, in addition to the difference in hue, the difference in brightness and vividness directly affects the color density value of each RGB color. Therefore, when the RGB image pickup data D0 is processed, a plurality of regions exist in the shaded portion based on the difference in the color density value due to the brightness and vividness even if there is no remarkable difference in hue. As a result, there is a possibility that the emulsion film 2 is not properly extracted. On the other hand, by targeting the HSV imaging data D1 in which the lightness component and the saturation component are separated from the hue component, the influence of such a difference in brightness and vividness is excluded. As a result, although it is visually recognized as a region with shading in the captured image IM1, it is originally a single hue value, or at most within a predetermined hue value centered on the hue value. It is possible to more accurately extract the formed region of the emulsion film 2, which only causes a difference in hue.

The extraction standard setting processing unit 180 is responsible for a process (extraction standard setting process) for setting a standard (extraction standard) used when extracting a predetermined region from an image of a measurement object. The extraction standard setting processing unit 180 includes a range selection unit 181 and a hue range specifying unit 182 for executing the extraction standard setting processing.

The range selection unit 181 indicates that the operator of the measuring device 100 selects (designates an area) an arbitrary pixel range of the image through the operation unit 133 while visually recognizing a predetermined image displayed on the display unit 132. It is a possible part. The operator of the measuring device 100 can perform the selection process while confirming the range to be selected in the captured image IM1 displayed on the display unit 132. By the action of the range selection unit 181 the operator selects a pixel range (small area) to be used for setting the extraction standard from the image while referring to the image of the measurement target displayed on the display unit 132. Can be done. In such a case, the image of the measurement object displayed on the display unit 132 may be based on the RGB imaging data D0, but may also be based on the HSV imaging data D1.

The method of selecting a specific area may be appropriately selected according to the type of the specific component (mouse, keyboard, touch panel, digitizer, etc.) of the operation unit 133. For example, a click operation and a drag operation by a mouse included in the operation unit 133, a key operation by a cursor key or the like of a keyboard, a direct writing on a touch panel by a stylus pen or the like, and the like are exemplified. From the viewpoint of convenience and versatility, the use of a mouse is preferable.

Further, the shape of the selected region may be any rectangle, polygon, or circle, but may be irregular.

The hue range specifying unit 182 specifies the hue values of individual pixels included in the region selected by the operator based on the HSV imaging data D1, and describes the extracted hue value range as the extraction standard. Generate data D2. The generated extraction reference data D2 is used for extraction processing in the region extraction processing unit 190.

The region extraction processing unit 190 is responsible for extracting a region satisfying the extraction criteria described in the extraction reference data D2 from the image given by the HSV imaging data D1. The region extraction processing unit 190 includes a corresponding pixel specifying unit 191 and a hue replacement processing unit 192 for these processes.

The corresponding pixel specifying unit 191 specifies a pixel whose hue value corresponds to the pixel range described as the extraction reference in the extraction reference data D2 among all the pixels included in the HSV imaging data D1. The range occupied by the image given by the HSV imaging data D1 by all the pixels specified by the processing is the region extracted by the extraction processing.

The hue replacement processing unit 192 generates the processing target data D3 by replacing the hue value of the pixel specified by the corresponding pixel identification unit 191 with a predetermined hue value (preferably one hue value).

The measurement processing unit 200 is a portion responsible for executing various measurement processes targeting an arbitrary portion of the image represented by the processing target data D3. The specific content of the measurement process varies depending on the formation mode of the emulsion film 2 and the opening 3 in each screen printing plate 10 (for example, measurement of the width of the opening 3, measurement of the diameter of the land, etc.). , Various known methods can be applied to the execution of the measurement process.

The measurement process may be an embodiment in which the operator measures the size of an arbitrary portion of the image while operating the operation unit 133 in a state where the display unit 132 displays the image represented by the processing target data D3. The mode may be such that it is automatically processed by the measurement processing program stored in the storage unit 134 in advance.

The measurement processing unit 200 generates the result of the measurement processing as the measurement data D4.

<Extraction processing of the formed region of the emulsion film>
Next, the image processing of the screen printing plate 10 for the captured image of the emulsion film 2 including the opening 3 executed in the measuring device 100 will be described. Specifically, the image processing is a processing for extracting a formation region of the emulsion film 2 from the captured image. FIG. 4 is a diagram showing the procedure of the image processing.

First, the screen printing plate 10 to be processed is placed on the stage 101 (step S1). At the same time, the screen printing plate 10 is fixed in an appropriate manner.

Subsequently, by operating the movement mechanism 100M under the control of the movement control unit 160, the image pickup unit 110 and the illumination unit 120 are positioned so that the target range of the extraction process in the emulsion film 2 is included in the image pickup range. .. Then, the screen printing plate 10 is imaged by the camera body 111 in a state where the lighting unit 120 irradiates the illumination light in a predetermined mode based on the control by the lighting control unit 150 (step S2).

The image pickup is performed by the camera body 111 after the optical system 113 in the lens barrel 112 is appropriately adjusted under the control of the image pickup processing unit 140. The RGB imaging data D0 generated by imaging is passed from the imaging processing unit 140 to the color space conversion unit 170. It is assumed that the captured image IM1 shown in FIG. 3 is an example of the captured image represented by the RGB captured data D0.

The color space conversion unit 170 sets the color density value of each pixel described based on the RGB color coordinate system in the RGB image pickup data D0 giving the image captured image IM1 shown in FIG. 3 in HSV color coordinates according to the above-mentioned known relational expression. It is converted into a description based on the system and HSV imaging data D1 is generated (step S3). Since the content of the image represented by the HSV imaging data D1 is substantially the same as that of the captured image IM1, the captured image IM1 shown in FIG. 3 may be referred to as an image represented by the HSV imaging data D1 for convenience. , Should be handled.

After the HSV imaging data D1 is generated, the extraction standard setting processing unit 180 acts to set an extraction standard for extracting the formation region of the emulsion film 2 from the image of the screen printing plate 10. The standard setting process is performed.

First, the range selection unit 181 displays the image pickup image IM1 on the display unit 132 based on the RGB image pickup data D0 or the HSV image pickup data D1, and the operator of the measuring device 100 operates the operation unit 133 to display the image on the display unit 132. A state in which an arbitrary pixel range of the captured image IM1 can be selected is realized.

When such a state is realized, the operator uses the operation unit 133 to visually recognize the captured image IM1 displayed on the display unit 132, and is a reference for extracting a region from the formed region of the emulsion film 2 in the captured image IM1. An operation for selecting a pixel range (small area) used for setting is performed (step S4). FIG. 5 is a diagram showing a captured image IM2 when a rectangular region RE is selected by an operator drag operation with a mouse (not shown) provided as an operation unit 133.

Conceptually or conceptually speaking, the range selection by the operator is a process of causing the measuring device 100 to grasp a representative example or a sample of an image of a region to be extracted (emulsion film 2 in the present embodiment) in the captured image IM1. It can be said that. Therefore, the operator selects at least a pixel range (small area) that he / she visually recognizes as being contained in the emulsion film 2 in the captured image IM1. Preferably, the operator selects a pixel range (small area) that he / she determines that the typical appearance of the emulsion film 2 is preferably exhibited.

In the case of the captured image IM1 shown in FIG. 3, in the image of the emulsion film 2, the shading consisting of the portion through which the screen mesh 4 existing in a grid pattern directly underneath is transmitted and the portion without transmission is repeated. Therefore, it is preferable that the operator selects the range so as to include the repetition of the shade to a considerable extent. The rectangular region RE shown in FIG. 5 is an example in which each portion is sufficiently included. Of course, a wider range may be selected. Alternatively, a plurality of rectangular regions RE separated from each other may be selected. On the other hand, it is not preferable to select a range that includes the opening 3.

When the range is selected by the operator, in response to this, the hue range specifying unit 182 specifies the hue range occupied by the hue values of all the pixels belonging to the rectangular region RE based on the description content of the HSV imaging data D1. Then, the range is set as an extraction standard (step S5). At that time, the hue range specifying unit 182 generates the extraction standard data D2 that describes the set extraction standard.

Since the RGB image pickup data D0 and the HSV image pickup data D1 differ only in the description method of the color density value and have common pixels, the captured image IM1 targeted for range selection by the operator is the RGB image pickup data D0. Even in the case based on the above, as long as the pixels belonging to the selected range are specified, the hue range specifying unit 182 can specify the hue range without any problem.

FIG. 6 is a diagram showing a histogram HG1 for hue values of all pixels belonging to the rectangular region RE shown in FIG. However, in FIG. 6, a standardized hue value is used as the hue value. In the histogram HG1, the hue range ΔHa whose frequency is not 0 is set as the extraction reference. The fact that the histogram HG1 having a peak and having a predetermined spread in the hue range as shown in FIG. 6 means that the hue in the rectangular region RE is not single, but mainly the hue value at the center of the peak. , Means that it is within the range of the predetermined hue value.

Depending on the emulsion film 2, a plurality of peaks may appear, and further, the hue range ΔHa having a non-zero power may be divided into a plurality of different ranges. In that case, all of these plurality of ranges serve as extraction criteria.

When the extraction reference data D2 is generated, the region extraction processing unit 190 subsequently identifies the pixels corresponding to the extraction criteria described in the extraction reference data D2, and replaces the hue values of the specified pixels. Is done.

First, the corresponding pixel specifying unit 191 refers to the description content of the HSV image pickup data D1 and identifies a pixel belonging to the hue range in which the hue value is set as the extraction reference among all the pixels constituting the image pickup image IM1 (. Step S6).

Subsequently, the hue replacement processing unit 192 can consider that the hue value of the pixel specified by the corresponding pixel identification unit 191 described in the HSV image pickup data D1 does not substantially appear in the image pickup image IM1 or It is replaced with a certain (preferably one) hue value that is not included in the hue values of the emulsion film 2 and the screen mesh 4 (step S7). Further, the region extraction processing unit 190 generates the processing target data D3 based on the result of the replacement by the hue replacement processing unit 192 (step S8).

FIG. 7 is a diagram illustrating a binary image IM3a showing a range occupied by a pixel specified by the corresponding pixel specifying unit 191 in black and an image IM3b after replacement by the hue replacement processing unit 192. FIG. 8 is a diagram showing a histogram HG2 for all the pixels constituting the captured image IM1 before replacement. However, also in FIG. 8, a standardized hue value is used as the hue value.

In the binary image IM3a shown in FIG. 7A, the region 2bk occupied by all the pixels specified by the corresponding pixel specifying unit 191 is shown in black. The binarization is for convenience of explanation, and it is not essential to make the region 2bk black. Further, in FIG. 7B, the hue value (or standardized hue value) of the pixel belonging to the region 2bk is replaced with a hue value (or standardized hue value) of 0 by the hue replacement processing unit 192. ing. Although it cannot be seen from FIG. 7B due to the limitation shown in the figure, the hue value (or standardized hue value) of 0 corresponds to red.

Comparing FIGS. 3 and 7, the hue value of the pixel is 0 in the black region 2bk in the image IM3a shown in FIG. 7A and the hue IM3b shown in FIG. 7B (the hue value becomes 0). The (replaced) region 2α substantially coincides with the region where the emulsion film 2 is formed in the captured image IM1 shown in FIG.

This means that the region 2α of the image IM3b formed by the substitution of the hue value by the hue replacement processing unit 192 corresponds to the formation region of the emulsion film 2 extracted from the captured image IM1. As a result, it means that the formation region of the emulsion film 2 is extracted by specifying the pixels to be included in the region 2α by the corresponding pixel specifying unit 191. Therefore, the region 2α is also referred to as an emulsion film equivalent region 2α.

Further, in the histogram HG2 shown in FIG. 8, while there are pixels having powers in a range other than the hue range ΔHa shown in FIG. 6, the normalized hue value different from any hue range is at least 0 to 40. In the hue range ΔHb, the frequency is substantially 0. Therefore, even if the hue value selected from the hue range ΔHb is adopted for the substitution by the hue replacement processing unit 192, the hue value other than the formation region of the emulsion film 2 is not affected. It is based on this point that the hue value of 0 is adopted when the image IM3b shown in FIG. 7B is obtained. That is, the replacement of the hue value by the hue replacement processing unit 192 also has an effect as a highlighting that enhances the distinctiveness of the emulsion film corresponding region 2α extracted as the formation region of the emulsion film 2 from other regions.

The above means that the formation region of the emulsion film 2 in the screen printing plate 10 is suitably extracted as the emulsion film corresponding region 2α by the extraction process based on the extraction reference data D2 in the region extraction processing unit 190. ..

It should be noted that the above-mentioned substitution with a hue value of 0 is merely an example. The specific value of the hue value used for the actual replacement in the hue replacement processing unit 192 may be appropriately set in advance by the operator of the measuring device 100 based on the captured image IM1, or is described in the HSV imaging data D1. It may be an embodiment set by the hue replacement processing unit 192 based on the hue value of each individual pixel. Further, it may be replaced with white or black, which has no tint. However, preferably, the hue value used for substitution is selected so that the emulsion film equivalent region 2α can be easily identified and the boundary portion between the emulsion film equivalent region 2α and the opening 3 can be easily identified in the measurement process in the subsequent step. To.

The data D3 to be processed may be the data itself described in HSV color coordinates or RGB color coordinates that gives the image IM3b after the replacement processing by the hue replacement processing unit 192, or is executed by the measurement processing unit 200. Depending on the content of the measurement, the data may be further processed. Examples of the latter include data obtained by extracting the edge portion of the emulsion film corresponding region 2α from the image IM3b.

Further, in the image IM3b, the formation region of the emulsion film 2 is replaced with the emulsion film corresponding region 2α, but instead, the region other than the formation region of the emulsion film 2 has a single hue value or white or black. May be replaced with.

Alternatively, data that gives a binarized image in which the region 2bk occupied by the extracted pixels is shown in black as in the image IM3a shown in FIG. 7A may be the processing target data D3.

Alternatively, as an embodiment for easily visually distinguishing the formed region of the extracted emulsion film 2, various methods may be applied instead of the substitution processing by the hue substitution processing unit 192. For example, a highlighting method may be applied, such as a mode in which only the boundary portion (edge portion) of the region is shown by a specific hue value, or a mode in which the boundary portion is blinked and displayed on the display unit 132. ..

Alternatively, a layer structure is given to the processing target data D3 so that the image of the screen printing plate 10 before replacement and the image of the emulsion film equivalent region 2α or the processed image thereof are expressed by different layers. It may be. In such a case, the measurement process may be performed on the latter layer.

The result of the measurement processing performed by the measurement processing unit 200 is output as the measurement data D4.

Since the emulsion film corresponding region 2α is a region consisting of a single hue value, microscopic irregularities are present due to the presence of the screen mesh 4, even if the region is the same as the original emulsion film 2 forming region. Even so, the shape can be specified, for example, the edge portion can be specified easily and surely.

Similar to the histogram HG1 for the hue values of all the pixels belonging to the rectangular region RE of FIG. 5 selected by the operator shown in FIG. 6, all the images IM1 before replacement shown in FIG. 8 are configured. Also in the histogram HG2 for pixels, there is a peak in the hue range ΔHa. Therefore, at first glance, it seems that the hue range used for the reference setting for region extraction can be specified from the histogram of the captured image IM1 as shown in FIG. 8 without intentionally selecting the range by the operator. However, in this case, it may not always be clear which hue value is derived from the image of the emulsion film 2. For example, in the case of the histogram HG2 shown in FIG. 8, the frequency is not 0 even in the range adjacent to the hue range ΔHa on the low hue value side (specifically, the range in which the hue value is about 170 to 180). Therefore, it is difficult to accurately determine at least the lower limit of the hue range corresponding to the image of the emulsion film 2 from the histogram HG2 shown in FIG. At best, it can be set to an appropriate value while leaving ambiguity.

For example, when the operator sets it artificially, there is a high possibility that individual differences will occur. Depending on the setting method, it is possible that the emulsion film corresponding region 2α finally specified does not necessarily sufficiently correspond to the actual formation range of the emulsion film 2. In addition, it is possible to eliminate artificial variation factors by using statistical processing methods such as discriminant analysis methods, but when the actual image is not always reflected and accurate region extraction cannot be performed. There is also.

The method according to the present embodiment is preferable in that the inaccuracy of region extraction due to such ambiguity is eliminated.

On the other hand, in the case of the method according to the present embodiment, the range selection by the operator for setting the extraction standard has arbitrary shapes, sizes, areas, and the like. However, as long as the selection from the formation range of the emulsion film 2 is made, the histogram based on the pixels included in the selected pixel range (small area) is different from the histogram HG1 shown in FIG. 6 due to the magnitude of the frequency. Even if there is, it is unlikely that a substantial difference will occur in the non-zero hue range ΔHa. Moreover, it is not necessary to select the entire formation range of the emulsion film 2. It is also possible to easily select a range with a mouse or the like.

Therefore, it can be said that the method according to the present embodiment has both convenience and redundancy in specifying the formation region of the emulsion film 2.

Further, FIG. 9 is a diagram showing an image IM1r obtained by extracting only the R component from the RGB image pickup data D0 that gives the image captured image IM1 and further binarizing it, which is shown for comparison. In the image IM1r, in addition to the emulsion film corresponding region 2β similar to the emulsion film corresponding region 2α in FIG. 7B being a black region, the image of the screen mesh 4 in the pattern opening 3 is also black. It is confirmed that it is. Although not shown, it has been confirmed that the same image as the image IM1r can be obtained for the G component and the B component. If only the formation region of the emulsion film 2 is to be extracted from the image, a process of separating the image of the screen mesh 4 from the emulsion film corresponding region 2β is indispensable.

This further means that it is not easy to directly extract only the formation region of the emulsion film 2 by using the image of the R component, the G component, or the B component of the RGB image pickup data D0. It is shown that the method according to the present embodiment is superior in terms of certainty (accuracy) and simplicity (easiness) of extraction.

As described above, according to the present embodiment, while visually recognizing the captured image of the emulsion film which is formed by coating on the screen mesh in the screen printing plate and has an opening in which the screen mesh is partially exposed. The entire region where the emulsion film is formed can be suitably extracted by a simple process of selecting a part of the range so as to include a portion where shading due to the presence or absence of transmission of the screen mesh is repeated.

In particular, by performing the processing based on the hue, the influence of the difference in brightness and vividness in the portion where the shading is repeated, which is a problem when the processing based on the color density value of each RGB color is performed, is eliminated. Accurate extraction is realized for the entire formed region of the emulsion film in which the shading is generated. Furthermore, by targeting the extracted formed region as a measurement target, it is possible to accurately measure the shape of each part of the screen printing plate, such as the shape of the opening.

<Modification example>
In the above-described embodiment, the hue range determined from the hue values of all the pixels belonging to the range selected by the operator is used as the extraction standard as it is, but the pixels whose hue values are near the lower limit and the upper limit of the hue range are Since it may be derived from noise instead of emulsion film, in order to eliminate the influence, a few percent (for example, 5%) of the lower and upper limits of the hue range is excluded when setting the extraction standard. There may be.

The image processing in the above-described embodiment targets the image of the screen printing plate and aims to improve the accuracy of measurement in the subsequent process, but this is not an essential aspect. The image processing can be applied to at least a general-purpose image when extracting a region in which a portion having a shade is repeated from a predetermined image.

<Application example of extraction processing>
The object of the series of image processing in the above-described embodiment is the image of the emulsion film 2 having repeated shading due to the presence or absence of transmission of the screen mesh 4, but the image processing is the object. Even if there is no such repetition in the shades generated in, in other words, even if the shades are not periodic, it may be suitably applicable.

Further, the extraction process does not necessarily have to be accompanied by a measurement process as a post-process, and it may be sufficient if a defect or an abnormal state can be detected and detected by visual inspection or pattern fitting.

FIG. 10 shows a case where the same image processing as that of the above-described embodiment is applied to the line pattern 51 formed by performing a predetermined pattern forming process on the main surface (base surface) of the metal plate 50 which is the base. It is a figure which shows the state. Examples of the pattern forming process include a film forming process and a coloring process.

FIG. 10A is an image captured image IM4a based on the RGB image pickup data D0 of the metal plate 50 by the image pickup unit 110. The captured image IM4a includes a part of the line pattern 51 (a portion bent at a right angle). The captured image IM4a is originally a color image, and in the color image, the metal plate 50 is visually recognized from cream to beige, and the line pattern 51 is visually recognized from blue to dark blue.

From the captured image IM4a, it is confirmed that the processing marks 52 are formed on the surface of the metal plate 50. The processing marks 52 are formed during processing (for example, during polishing) of the metal plate 50 prior to the formation of the line pattern 51, and are streaked in the vertical and horizontal directions in the drawing, with irregular (random) intervals and lengths. , And in thickness.

More specifically, a part of the processing mark 52 is also present in the base portion of the line pattern 51. This is particularly referred to as a base processing mark 52a. In the line pattern 51, due to the random presence of such surface processing marks 52a, the overall color is generally uniform blue to dark blue, but the portion where the color is slightly different is random. Is occurring in. In other words, irregular shades are visible.

FIG. 10B shows an extraction standard for such a captured image IM4a by performing the same color space conversion process as in the above-described embodiment and by selecting a partial range (small area) of the line pattern 51 by the operator. It is an image IM4b obtained by performing the setting process, the extraction process based on the set extraction standard, and the replacement process based on the extraction process result. In addition, unlike the captured image IM1 in which repeated shades are visually recognized, in the case of the captured image IM4a, the shades are irregular, so that the operator selects a partial range (small area) to include such shades to a considerable extent. As such, it is preferable to be done.

In the image IM4b, the hue value (or standardized hue value) of the pixel specified (extracted) by the corresponding pixel identification unit 191 becomes 0 hue value (or standardized) corresponding to red by the hue replacement processing unit 192. Hue value). Hereinafter, the region consisting of the pixels is referred to as the extracted region 51α. Regions other than the region to be extracted 51α are white.

The existing range of the extracted region 51α in the image IM4b generally matches the existing position of the line pattern 51 in the captured image IM4a shown in FIG. 10 (a). In particular, in FIG. 10B, there is a protruding portion in the range surrounded by the circle C in the extracted region 51α, which corresponds to the defective portion 51a protruding from the line pattern 51 in the captured image IM4a. Is.

This means that it is possible to inspect the presence or absence of defects in the line pattern 51 by observing the image obtained by applying the series of image processing according to the above-described embodiment. is doing.

That is, by applying the series of image processing in the above-described embodiment, not only the region where shading is periodically generated but also the region where irregular shading is generated such as the line pattern 51. , Can be suitably extracted.

If an attempt is made to extract (specify) the line pattern 51 by subjecting the captured image IM4a shown in FIG. 10A to a binarization process as described in Patent Document 1, the metal plate 50 itself It becomes difficult to distinguish between the processing marks 52 existing in the above and the line pattern 51 formed by the pattern forming process. On the other hand, according to the image processing method using hue as in the above-mentioned application example, the pattern forming process is performed on the metal plate 50 even when the processing marks 52 are present. The region (line pattern 51) formed thereby can be reliably extracted by a simple method.

Based on the above, the series of image processing in the above-described embodiment targets an image in which a predetermined pattern is formed on the base surface, which appears as "periodic or irregular shading" when imaged. In some cases, it can be said to be suitable. Such include mesh screens, striations, uneven surfaces, glossy surfaces, and the like. In particular, when it is difficult to distinguish between the light and darkness derived from the background and the pattern by the conventional binarization processing, the case where the pattern formation processing is applied to the underground of the type in which light and darkness are mixed when imaged is targeted. It can be said that there is an advantage.

1 Plate frame 2α Emulsion film equivalent area 2 Emulsion film 3 Opening (pattern opening)
4 Screen mesh 4x Weft 4y Warp 10 Screen printing plate 50 Metal plate 51 Line pattern 52 Processing mark 52a Base processing mark 100 Measuring device 100M Moving mechanism 101 Stage 110 Imaging unit 111 Camera body 112 Lens barrel 113 Optical system 120 Lighting unit 121 Coaxial lighting 122 Ring lighting 123 Transmission lighting 130 Processing device

Claims (6)

An image processing device that extracts a predetermined area that matches the existence position of the line pattern from the image to be processed including the image of the line pattern.
A display unit on which the image to be processed is displayed and
An operation unit that can perform a predetermined operation on the image to be processed, and
An extraction standard setting processing unit that sets an extraction standard when extracting the predetermined area, and
An area extraction processing unit that extracts the predetermined area from the image to be processed based on the extraction standard, and a region extraction processing unit.
Equipped with
The predetermined region is a region where the hue is within a predetermined hue value range, and is a region having irregular shading due to processing marks existing in the base portion of the line pattern .
The extraction standard setting processing unit is
A range selection unit that enables an operator to select an arbitrary pixel range from the processing target image by operating the operation unit while visually recognizing the processing target image displayed on the display unit.
Hue range specification that specifies the range of hue values of pixels belonging to the arbitrary pixel range selected by the operation of the operation unit by the operator based on the data of the processing target image described in the HSV color coordinate system. Department and
A small region including the hue from the region that the operator visually recognizes as being included in the predetermined region in the processed image, which is made by the operator as the selection of the arbitrary pixel range. In response to the selection of, the range in which the entire range of the hue values of the pixels included in the small area specified by the hue range specifying unit is removed or the predetermined ratios of the preset upper and lower limits are removed. , Set as the extraction standard,
The area extraction processing unit is
A pixel identification unit that identifies pixels whose hue value corresponds to the extraction criterion in the image to be processed.
And extracts a region consisting of all the pixels specified by the pixel specifying unit as the predetermined region.
An image processing device characterized by this. The image processing apparatus according to claim 1.
An imaging processing unit that captures a predetermined imaging target and generates imaging data described in the RGB color coordinate system.
A color space conversion unit capable of converting the data of the image to be processed described in the RGB color coordinate system into the data of the HSV color coordinate system, and
Further prepare
The data of the processing target image described in the HSV color coordinate system is obtained by converting the imaging data obtained by imaging a predetermined imaging target with the imaging processing unit by the color space conversion unit. Be,
An image processing device characterized by this. The image processing apparatus according to claim 1 or 2 .
The area extraction processing unit
A hue replacement processing unit that replaces the hue value of a pixel specified by the pixel identification unit with a hue value different from the hue value that gives an image of the predetermined region in the image to be processed.
An image processing apparatus characterized by further having. The image processing apparatus according to any one of claims 1 to 3 .
The entire or end portion of the predetermined region extracted by the region extraction processing unit is highlighted in the processing target image displayed on the display unit.
An image processing device characterized by this. An image processing method for extracting a predetermined region matching the existence position of the line pattern from a processing target image obtained by imaging a predetermined imaging target including an image of a line pattern.
An extraction standard setting step for setting an extraction standard when extracting the predetermined area, and
A region extraction step of extracting the predetermined region from the image to be processed based on the extraction standard, and a region extraction step.
Equipped with
The predetermined imaging target has a predetermined pattern formed on the base surface.
The predetermined region is a region included in the image of the predetermined pattern in the image to be processed and whose hue is within the range of the predetermined hue value, and is a processing mark existing in the base portion of the line pattern. Areas with irregular hues due to
The extraction standard setting process
By operating the predetermined operation means while visually recognizing the processing target image displayed on the predetermined display means, the operator can visually recognize that the processing target image is included in the predetermined area. A range selection step of selecting an arbitrary small area including the shade, and
A hue range specifying step of specifying a range of hue values of pixels belonging to the arbitrary small area based on the data of the processed image described in the HSV color coordinate system.
The range obtained by removing the entire range of the hue values of the pixels belonging to the arbitrary small region or the predetermined ratios of the upper and lower limits set in advance , which are specified in the hue range specifying step. Is set as the extraction standard,
The region extraction step
A pixel identification step of specifying a pixel whose hue value corresponds to the extraction criterion in the image to be processed.
And extracts a region consisting of all the pixels specified in the pixel specifying step as the predetermined region.
An image processing method characterized by that. The image processing method according to claim 5 .
An imaging processing step of imaging the predetermined imaging target and generating imaging data described in the RGB color coordinate system, and
A color space conversion step capable of converting the data of the image to be processed described in the RGB color coordinate system into the data of the HSV color coordinate system, and
Further prepare
The data of the processing target image described in the HSV color coordinate system is obtained by converting the imaging data obtained by imaging a predetermined imaging object in the imaging processing step in the color space conversion step. can get,
An image processing method characterized by that.

Images