JP2010211022A - Method for adjusting focal position of image recognition optical system, image recognition optical system, and head alignment device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily adjust a focal position, and to accurately obtain an image in focus of an object to be recognized. <P>SOLUTION: The method for adjusting the focal position of an image recognition optical system includes a measuring step of measuring a plurality of focus evaluated values for indexing sharpness of the image in focus, relative to the moving amount of a focus lens 86, while the focal position is moved forward, from a front side of an assumed proper focal position; a decrease-evaluated value detecting step of detecting a decrease focus evaluated value which begins to decrease, after passing through a peak out of the plurality of measured focus evaluated values; an effective evaluated value extracting step of extracting a plurality of focus evaluated values, equal to or above the decrease focus evaluated value out of the plurality of measured focus evaluated values as effective evaluated values; and a focal position adjusting step of adjusting the focal position, by forming an approximate curve from the plurality of effective evaluated values and moving the focus lens 86 by the moving amount of the peak. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a focus position adjusting method for an image recognition optical system, an image recognition optical system, and a head alignment apparatus that adjust a focus position by moving a focus lens of the image recognition optical system in order to obtain a focused image of a recognition object. It is about.

  Conventionally, as a focus position adjustment method of this type of image recognition optical system, a method of adjusting a focus position using a focus evaluation value (focus evaluation value) based on an image recognition result is known (see Patent Document 1). . In this focus position adjustment method, image recognition is performed at a large number of positions within a search range, focus evaluation values at each position are obtained, and then the focus position is adjusted to a position with the highest focus evaluation value. .

JP 2007-212724 A

  However, in such a focus position adjustment method, since the accuracy with which a focused image is obtained changes in proportion to the image recognition interval within the search range, it is necessary to perform image recognition at fine intervals. Therefore, there is a problem that the adjustment of the focal position becomes complicated. On the other hand, a method of generating an approximate curve using each focus evaluation value as a sample and widening the image recognition interval and adjusting the focal position to the peak position of the approximate curve is also conceivable. Since the focus evaluation value at a distant position is also used as a sample, there is a problem that an approximate curve near the target peak position cannot be obtained with high accuracy, and as a result, a focused image cannot be obtained with high accuracy. Further, since the number of focus evaluation values that are samples of the approximate curve is biased before and after the peak position, the approximate curve before and after the peak position cannot be generated under the same conditions. Therefore, there is a problem that the peak position of the approximate curve is shifted back and forth, and a focused image cannot be obtained with high accuracy.

  The present invention provides a focus position adjusting method for an image recognition optical system, an image recognition optical system, and a head alignment device that can easily adjust a focus position and can obtain a focused image of a recognition object with high accuracy. The issue is to provide.

  The focus position adjustment method of the image recognition optical system according to the present invention adjusts the focus position by moving the focus lens of the image recognition optical system in order to obtain a focused image of the recognition object. A focus evaluation value that indicates the sharpness of the focused image with respect to the moving amount of the focus lens while moving the focus position from the front of the assumed proper focus position by moving the focus lens. Decrease evaluation value detection that detects a reduced focus evaluation value, which is a focus evaluation value that starts to decrease through a peak with the amount of movement of the focal lens, from a measurement process that measures multiple locations and a plurality of measured focus evaluation values An effective evaluation value extracting step of extracting, as effective evaluation values, a plurality of focus evaluation values that are equal to or greater than the reduced focus evaluation value among the plurality of measured focus evaluation values; Generating an approximate curve from a plurality of effective evaluation value, only the amount of movement of the peak, by moving the focus lens, characterized by comprising a focus position adjusting step of adjusting the focal position.

  The image recognition optical system of the present invention is an image recognition optical system capable of adjusting a focal position by moving a focus lens of an image recognition optical system in order to obtain a focused image of a recognition object. Measuring means for measuring a plurality of focus evaluation values for measuring the sharpness of the focused image with respect to the moving amount of the focusing lens while moving the focal position from the front of the assumed appropriate focal position to the front. Among the plurality of focus evaluation values, a reduced evaluation value detection means for detecting a reduced focus evaluation value, which is a focus evaluation value that starts to decrease through a peak with the amount of movement of the focus lens, and a plurality of measured focus evaluations Among these values, an effective evaluation value extracting means for extracting a plurality of focus evaluation values equal to or greater than the reduced focus evaluation value as effective evaluation values, and generating an approximate curve from the plurality of effective evaluation values. And, by moving the moving amount by a focal lens of the peak, characterized by comprising a focus position adjusting means for adjusting the focal position.

  According to these configurations, a focus evaluation value that is equal to or greater than the reduced focus evaluation value that starts to decrease after the peak is used as an effective evaluation value (sample), and the focus evaluation value in the vicinity of the peak position is obtained by generating an approximate curve. Since an approximate curve is generated as a sample, an approximate curve near the target peak position can be obtained with high accuracy. In addition, since the sample ranges before and after the peak position are the same and the number of samples before and after the peak position is not biased, the peak position of the approximate curve does not shift. Therefore, a focused image to be recognized can be obtained with high accuracy. In addition, since it is not necessary to perform image recognition at fine intervals, the focus position can be easily adjusted.

  The head alignment apparatus of the present invention is a head alignment apparatus that mutually positions a plurality of inkjet heads having alignment marks on the nozzle surface, and has an alignment having a plurality of reference marks corresponding to the alignment marks of the plurality of inkjet heads. Mask, mask fixing jig for positioning and fixing the alignment mask on the set table, alignment marks for a plurality of inkjet heads set in a pre-aligned state on the mask fixing jig, and alignment mask reference The image recognition optical system for simultaneously recognizing images of the mark from below, and position correction means for correcting the position of each of the plurality of inkjet heads based on the recognition result of the image recognition optical system. .

  According to this configuration, the inkjet head can be positioned easily and accurately by using the image recognition optical system that can adjust the focal position easily and accurately.

  In this case, the image recognition optical system is preferably composed of a bifocal microscope including a first optical system for recognizing the alignment mark and a second optical system for recognizing the reference mark.

  According to this configuration, by using a bifocal microscope that recognizes images of the alignment mark and the reference mark, it is possible to focus on each of the marks, so that a focused image of each mark can be obtained simultaneously with high accuracy. And alignment can be performed easily and accurately.

It is a disassembled perspective view of the head unit which aligns with the head alignment apparatus which concerns on this embodiment. It is a disassembled perspective view of the inkjet head mounted in a head unit. It is sectional drawing of an inkjet head. It is sectional drawing of the head alignment apparatus which concerns on this embodiment. FIG. 5 is a cross-sectional view taken along line AA in FIG. 4 of the head alignment apparatus according to the present embodiment. It is the flowchart which showed the focus position adjustment process of each optical system. It is explanatory drawing for demonstrating appropriate position determination operation | movement.

  Hereinafter, a head alignment apparatus according to an embodiment of the present invention will be described with reference to the accompanying drawings. This head alignment apparatus is for forming a head unit composed of a plurality of inkjet heads, and positions the plurality of inkjet heads relative to each other. In addition, this head alignment apparatus is characterized by using a bifocal microscope that can be focused accurately and easily. Prior to the description of the head alignment apparatus, an inkjet head to be subjected to the alignment and a head unit equipped with a plurality of the inkjet heads will be described.

  As shown in FIG. 1, the head unit 1 includes a cartridge case 11 (only the bottom plate is shown) that holds and holds a plurality of (four colors (black, red, green, blue)) ink cartridges that are not shown. 4) inkjet heads 12, a fixed plate 13 to which each inkjet head 12 is bonded and fixed, and a cover head 14 that covers the inkjet head 12 fixed to the fixed plate 13.

  In the cartridge case 11, a plurality of ink supply needles 11 a (with filters) to which ink cartridges of respective colors as ink supply means are respectively attached are arranged in a line at equal intervals. Each of the ink supply needles 11a has a plurality of minute ink communication paths 11b, and each ink communication path 11b communicates with a pair of ink inlets 19a of a head case 19 described later.

  As shown in FIG. 2 and FIG. 3, the inkjet head 12 includes a nozzle plate 15 having a plurality of discharge nozzles 21, and a flow path forming substrate 16 in which a pressure chamber 16 a communicating with each discharge nozzle 21 is formed. A protective substrate 17 that protects the piezoelectric element 23, a seal substrate 18 that seals the flow path forming substrate 16 and the common chamber 16 b communicating with the protective substrate 17, and a head case 19 that forms the exterior of the inkjet head 12. The layers are laminated in order from the side.

  The nozzle plate 15 is formed of stainless steel or the like, and is bonded to one surface (lower surface) of the flow path forming substrate 16 with an adhesive or the like. In addition, a plurality (180) of discharge nozzles 21 are perforated at equal pitches on the nozzle surface NF of the nozzle plate 15 to form one nozzle row NL. Two nozzle rows NL are formed in parallel on the nozzle surface NF, and the plurality of discharge nozzles 21 on the nozzle surface NF of each row are arranged with a half-pitch deviation from each other (in FIG. 2, No deviation between the nozzle rows NL is shown).

  In the flow path forming substrate 16, a plurality of pressure chambers 16 a partitioned by a plurality of partition walls are formed in two rows in the width direction so as to correspond to the plurality of discharge nozzles 21. In addition, a pair of common chambers 16b for storing ink supplied from the ink cartridges of the respective colors are formed along the long side in a region on the outer side in the short side direction of the flow path forming substrate 16, and each pressure chamber 16a and The common chamber 16b communicates with each other via a supply path 16c. The common chamber 16 b communicates with the protective substrate 17 and receives ink supplied from the ink cartridges of each color via an ink introduction port 19 a provided in the head case 19.

  The protective substrate 17 is connected to the flow path forming substrate 16 through an elastic film 17a. The number of piezoelectric elements 23 corresponding to each pressure chamber 16a is formed on the elastic film 17a, and a mechanism portion 17b for housing each piezoelectric element 23 is provided in a recessed manner. A drive IC 24 for driving each piezoelectric element 23 is provided on the protective substrate 17, and each terminal of the drive IC 24 is connected to each piezoelectric element 23 via a bonding wire or the like. Each terminal of the drive IC 24 is connected to the outside via an external wiring 25 (flexible flat cable) and receives various signals such as a print signal.

  The head case 19 is connected to the protective substrate 17 via the seal substrate 18, and the common chamber 16 b is sealed only by the seal substrate 18. The head case 19 has a pair of ink inlets 19a that serve as ink inlets, and is open at substantially the center in the long side direction. Further, an IC holding portion 19b through which the external wiring 25 connected to the driving IC 24 is inserted is formed in the center portion of the head case 19 in the thickness direction.

  Such an inkjet head 12 takes in ink from the ink cartridge through the ink communication path 11b (see FIG. 1) and the ink introduction port 19a, and fills the ink from the common chamber 16b to each discharge nozzle 21 with ink. In this state, by applying a voltage to the piezoelectric element 23 to deform the elastic film 17a, the functional liquid is introduced from the common chamber 16b using the volume change of the pressure chamber 16a, and ink droplets are ejected from the ejection nozzle 21. Discharge.

  Next, as shown in FIG. 1, the fixing plate 13 is a flat plate made of stainless steel or the like, and is a common fixing member that is bonded to the nozzle surfaces NF of the four inkjet heads 12. The heads 12 are held in a line at equal intervals in the short side direction. The fixed plate 13 defines an exposed opening 13a that exposes the plurality of ejection nozzles 21 of each inkjet head 12, and a frame-shaped joint 13b that defines the exposed opening 13a and is joined to the periphery of the nozzle surface NF of the inkjet head 12. (Consisting of a four-frame frame and three crosspieces). The positioning of each inkjet head 12 with respect to the fixed plate 13 is performed using a head alignment device 2 described later. The fixing plate 13 and the nozzle plate 15 are joined using, for example, a thermosetting epoxy adhesive, an ultraviolet curable adhesive, or the like. The head unit 1 according to the present embodiment includes four inkjet heads 12, but the number is arbitrary.

  The cover head 14 is a box-like member having a frame-like bottom plate made of stainless steel or the like, and is fixed to the cartridge case 11 so as to cover the four inkjet heads 12 fixed to the fixing plate 13. . Further, the cover head 14 is provided with four cover openings 14 a corresponding to the four exposed openings 13 a of the fixed plate 13.

  Next, the head alignment device 2 for positioning the plurality of inkjet heads 12 on the fixed plate 13 will be described with reference to FIGS. 4 and 5. The head alignment apparatus 2 integrally presses the inkjet head 12 toward the fixing plate 13 side with an alignment jig (mask fixing jig) 31 on which the inkjet head 12 to be aligned is placed, and the alignment jig 31. A pair of bifocal microscopes having a pressing means 32, a moving stage (set table) 33 for setting the alignment jig 31 and moving it linearly, and an optical system for imaging the inkjet head 12 from below the moving stage 33 (Image recognition optical system) 34, an alignment mechanism (position correction means) 35 for performing predetermined alignment of the inkjet head 12 via the chuck 63c, a control for controlling the moving stage 33, each bifocal microscope 34 and the alignment mechanism 35 Device (PC (Personal Comp Comprises ter) or the like (not shown)).

  The alignment jig 31 includes an alignment mask 36 provided with a plurality of reference marks 42 that serve as a reference for positioning each inkjet head 12, a base jig 37 for fixing the alignment mask 36 to the moving stage 33, and a base jig. And a spacer jig 38 for holding the fixing plate 13 disposed on the tool 37. The base jig 37 and the spacer jig 38 are made of stainless steel or the like.

  The alignment mask 36 is made of a light-transmitting material such as glass (in the embodiment, quartz glass), and has a plurality of convex portions 41 having respective reference marks 42 formed at the tip portion. Each of the convex portions 41 includes a first nozzle and a 360th nozzle (hereinafter referred to as a reference) that are diagonally located on the outer side in the long side among the total 360 discharge nozzles 21 formed on the nozzle plate 15 of each inkjet head 12. It protrudes to a position corresponding to the nozzle 22 (see FIG. 2). That is, a pair of convex portions 41 protrudes for each inkjet head 12, and a total of eight convex portions 41 protrude from the alignment mask 36. And the position of each nozzle plate 15 (position of the inkjet head 12) on the fixed plate 13 is accurately positioned by matching the reference marks 42 and the reference nozzles 22 provided on the respective convex portions 41 with each other. Can be done. That is, each reference nozzle 22 is used as a positioning mark for the inkjet head 12.

  Here, each reference mark 42 is preferably formed at a height near the nozzle surface NF of the nozzle plate 15. This is because the optical path is greatly shifted due to the heat of a metal halide lamp or the like used when recognizing an image with each bifocal microscope 34, and a large error occurs in the actual position between each reference mark 42 and reference nozzle 22. This is to prevent it. Thereby, positioning accuracy improves.

  The base jig 37 is formed in a bottomless box shape, and is detachably disposed on the moving stage 33 so as to cover the alignment mask 36 placed on the moving stage 33. Further, eight base through holes 43 through which the convex portions 41 of the alignment mask 36 face are formed through the top plate of the base jig 37 in the thickness direction. Needless to say, the alignment mask 36 is positioned on the moving stage 33 by the base jig 37.

  The spacer jig 38 is a so-called suction plate and holds the fixed plate 13 by suction. The spacer jig 38 includes a plurality (five) of suction chambers 44 to which suction means such as a vacuum pump (not shown) is connected between the upper surface of the base jig 37 (see FIG. 4). . Each suction chamber 44 has openings (adsorption holes) arranged in a line at equal intervals in the long side direction at substantially the center in the short side direction of the spacer jig 38, and the fixing plate 13 (the frame-like joint portion 13b). It is designed to suck and hold the back side. The spacer jig 38 is positioned and fixed to the base jig 37, and the fixing plate 13 is positioned and fixed to the base jig 37. In other words, the fixing plate 13 and the alignment mask 36 are roughly positioned with respect to each other.

  Further, the spacer jig 38 is formed with eight spacer through holes 45 communicating with the respective base through holes 43 in the thickness direction. Thereby, each reference nozzle 22 of the nozzle plate 15 held on the fixed plate 13 recognizes an image by each bifocal microscope 34 from the bottom surface side of the alignment mask 36 through the spacer through hole 45 and the base through hole 43. Can be done.

  The pressing means 32 includes a frame portion 46 formed in a bottomless box shape so as to cover the inkjet head 12 set on the spacer jig 38, and provided on the upper surface of the frame portion 46. And four pressing portions 47 for pressing the plate 15) toward the fixed plate 13 side. A plurality of chuck openings 48 into which a pair of chucks 63c of the alignment mechanism 35 described later is inserted are formed below both side surfaces in the short side direction of the frame portion 46 (see FIG. 5). Each chuck opening 48 is opened at a position corresponding to each inkjet head 12 set on the spacer jig 38.

  Each pressing portion 47 is inserted into the frame portion 46 and provided with a pressing pin 51 provided so as to be able to advance and retreat in the vertical direction, and the pressing pin 51 is biased toward the inkjet head 12 by a biasing spring 52a and the pressing force is adjusted. An urging means 52 having a possible screw mechanism portion 52 b and a block-shaped pressing pad 53 disposed between the pressing pin 51 and the inkjet head 12 are configured. The urging means 52 can adjust the pressure with which the urging spring 52a presses the pressing pin 51 by the tightening amount of the screw mechanism portion 52b. Thus, the pressing force for pressing each inkjet head 12 against the fixed plate 13 can be adjusted.

  The moving stage 33 includes a stage portion 54 that sets the alignment jig 31 at a predetermined position, and a linear motor (not shown) that moves the stage portion 54 linearly (X-axis direction). In the stage portion 54, eight stage through holes 56 communicating with the spacer through hole 45 and the base through hole 43 are formed, and an optical path for imaging with each bifocal microscope 34 is secured.

  Each bifocal microscope 34 simultaneously reproduces two images with different focal points (image recognition) via two optical systems 71 and 72 (first optical system 71 and second optical system 72) sharing the optical axis L. A microscope optical system 73 constituting the first optical system 71 and the second optical system 72, and an image pickup unit that reproduces an image captured simultaneously via the first optical system 71 and the second optical system 72. 74 (CCD camera). The optical axis L is directed from the lower side of the moving stage 33 toward the reference mark 42 and the reference nozzle 22.

  The microscope optical system 73 includes an objective lens 76 that forms an optical axis L, a split beam splitter 77 that vertically reflects part of incident light from the objective lens 76 and transmits part of the incident light, and a split beam. A first mirror 78 that vertically reflects the reflected light of the splitter 77 and guides it to an integrated beam splitter 82, which will be described later, and a first focusing portion 79 disposed on the optical path from the split beam splitter 77 to the first mirror 78. And a second mirror 80 that vertically reflects the transmitted light of the split beam splitter 77 and guides it to the integrated beam splitter 82, and a second focus that is disposed on the optical path from the second mirror 80 to the integrated beam splitter 82. Unit 81 and an integrated beam splitter 82 that vertically reflects the light from the first mirror 78 and transmits the light from the second mirror and integrates them into the imaging unit 74. It is equipped with a. The first optical system 71 includes an objective lens 76, a split beam splitter 77, a first mirror 78, a first focusing unit 79, and an integrated beam splitter 82. On the other hand, the second optical system 72 includes an objective lens 76, a split beam splitter 77, a second mirror 80, a second focusing unit 81, and an integrated beam splitter 82.

  Each focusing unit 79, 81 includes a focus lens 86 disposed on the optical path, and a lens moving mechanism 87 that moves the focus lens 86 back and forth on the optical path. The focal positions of the optical systems 71 and 72 are adjusted by moving the focus lenses 86 of the focusing units 79 and 81 in the optical path direction. In each bifocal microscope 34, the focal position of the first optical system 71 is adjusted to the reference mark 42 (height) and the focal position of the second optical system 72 is adjusted to the reference nozzle 22 (height). Both in-focus and clear images (in-focus images) can be obtained on the imaging unit 74 at the same time. A predetermined alignment is performed by adjusting the position of the inkjet head 12 so that these images overlap. Note that the movement position (appropriate movement position) of the focus lens 86 when adjusting the focal position is determined by an appropriate position determination operation described later.

  The alignment mechanism 35 is vertically suspended from the apparatus frame 61, a moving mechanism 62 including a Y-axis stage 62a, an X-axis stage 62b, and a θ-axis stage 62c in order from the upper side supported by the apparatus frame 61, and the θ-axis stage 62c. And a gripping mechanism 63 that grips each inkjet head 12. The gripping mechanism 63 includes a pair of arms 63a suspended from the θ-axis stage 62c, a drive cylinder 63b disposed at the tip of each arm 63a, and a chuck 63c fixed to the tip of the piston rod of each drive cylinder 63b. By driving the pair of drive cylinders 63b, the pair of chucks 63c passes through the chuck openings 48 opened on both side surfaces of the frame portion 46, and the long side direction of one inkjet head 12 It abuts on both end faces of and grips it.

  Then, by driving the Y-axis stage 62a, the X-axis stage 62b, and the θ-axis stage 62c, the one inkjet head 12 that is pressed by the pressing means 32 with a predetermined pressing force is moved in the X-axis / Y-axis directions. And rotate around an axis orthogonal to the XY plane. That is, each inkjet head 12 is aligned by the alignment mechanism 35 in a state where the inkjet head 12 is pressed against the fixed plate 13 coated with an adhesive (non-cured state). The moving mechanism 62 is fixed in position with respect to the moving stage 33 and faces the individual ink jet heads 12 as the moving stage 33 moves in the X-axis direction in the figure.

  In the head alignment apparatus 2 as described above, the alignment jig 31 positioned and set on the moving stage 33 in accordance with a command from the control apparatus is moved horizontally by the moving stage 33 perpendicular to the optical axis L of each bifocal microscope 34. Move appropriately in the X-axis direction. Then, each reference nozzle 22 of each ink jet head 12 faces the optical axis L, and the alignment mechanism 35 is driven based on the image recognition result by the pair of bifocal microscopes 34, and each reference mark 42 and each reference nozzle are driven. 22 is matched. In this manner, the fixed plate 13 and the nozzle plate 15 of each inkjet head 12 are bonded with an adhesive while being positioned at a predetermined position. Note that the adhesive is applied to the side of the fixed plate 13 that is to be bonded to the nozzle plate 15. Further, the positioning of the fixing plate 13 and each inkjet head 12 performed in the head alignment device 2 is temporarily fixed. Thereafter, the alignment jig 31 is taken out from the head alignment device 2 and carried into another device. Then, the curing process of the adhesive is performed.

  Here, with reference to FIG. 6 and FIG. 7, the focus position adjustment process of each optical system 71 and 72 is demonstrated. This focus position adjustment process constitutes a so-called passive autofocus function, and in order to obtain a focused image of the recognition object (reference mark 42 and reference nozzle 22), the focus lens 86 is moved to focus. This is a process of adjusting the position. As shown in FIG. 6, the focus position adjustment process is an appropriate position determination operation (S1 to S1) for determining a movement position of the focus lens 86 (hereinafter referred to as an appropriate movement position) for obtaining a focused image of the recognized captured object. S5) and a focus position adjustment operation (S6) for adjusting the focus position by moving the focus lens 86 to the determined proper movement position. These are performed under the control of the control device.

  As shown in FIG. 6, in the proper position determination operation, first, a focus evaluation value (so-called focus) with respect to a movement amount from the reference position (origin) while the focus lens 86 is moved within the search range by the lens moving mechanism 87. (Value) is measured at a plurality of locations at predetermined intervals (S1) (measuring step and measuring means). The search range is a movement range of the focus lens 86 that moves the focus position within a range near the proper focus position. The proper focus position is a rough focus position where a focused image is assumed to be obtained, and the range in the vicinity of the proper focus position is a range from the front of the proper focus position to the front. The focus evaluation value is a value that indicates the sharpness of the focused image. The higher the focus evaluation value, the clearer the image that can be obtained.

  Each focus evaluation value is measured by reproducing the image of the recognition target object by the imaging unit 74 and obtaining the focus evaluation value based on the sharpness of the reproduction result. That is, while moving the focus lens 86, the image of the recognition object is reproduced at each movement amount (each movement position) from the reference position, and the focus evaluation value of the movement amount is obtained based on the reproduction result. . Thereby, each focus evaluation value for each movement amount is measured. As a result, a graph as shown in FIG. 7A is created.

  When each focus evaluation value for each movement amount is measured, a reduced focus evaluation value that starts to decrease beyond the peak with the movement amount of the focus lens 86 is detected from the measured focus evaluation values (S2). (Refer FIG.7 (b)) (decrease evaluation value detection process and reduction evaluation value means). Strictly speaking, a first focus evaluation value that is determined to have decreased beyond the peak as the movement amount increases is detected as a reduced focus evaluation value. That is, in other words, it is determined whether or not each of the plurality of measured focus evaluation values is a reduced focus evaluation value, and a focus evaluation value determined to be a reduced focus evaluation value is detected. For example, the focus evaluation value of the next movement amount of the movement amount having the maximum focus evaluation value among the plurality of focus evaluation values is detected as the reduced focus evaluation value. In consideration of the influence of the measurement error, the focus evaluation value of the movement amount having the maximum value, the movement amount of the second and third destinations may be used as the reduced focus evaluation value, or the movement of the three highest focus evaluation values is performed. The focus evaluation value of the next movement amount of the amount may be used as the decrease focus evaluation value.

  When the reduced focus evaluation value is detected, among the plurality of measured focus evaluation values, a plurality of focus evaluation values equal to or greater than the reduced focus evaluation value are extracted as effective evaluation values (S3) (see FIG. 7C) (valid Evaluation value extraction step and effective evaluation value extraction means). Thereafter, an approximate curve of the focus evaluation value with respect to the movement amount is generated based on the plurality of effective evaluation values (S4) (see FIG. 7D). When the approximate curve is generated, the position moved from the reference position by the amount of movement of the peak (vertex) of the approximate curve is determined as the proper movement position of the focus lens 86 (S5) (see FIG. 7E).

  When the proper movement position of the focus lens is determined, the focus position adjustment operation is performed. That is, the lens moving mechanism 87 moves the focal lens 86 to the determined proper movement position (S6) (focal position adjustment step and focal position adjustment means). As a result, the focal position is adjusted to a position where a focused image of the recognition object is obtained, and the present process ends.

  According to the above configuration, the focus evaluation value in the vicinity of the peak position is generated by generating an approximate curve using the focus evaluation value equal to or higher than the reduced focus evaluation value starting to decrease after the peak as an effective evaluation value (sample). Since an approximate curve is generated using the value as a sample, an approximate curve near the target peak position can be obtained with high accuracy. Further, since the number of samples before and after the peak position is not biased, the peak position of the approximate curve is not shifted. Therefore, a focused image to be recognized can be obtained with high accuracy. In addition, since it is not necessary to perform image recognition at fine intervals, the focus position can be easily adjusted.

  Further, by using an image recognition optical system (bifocal microscope 34) that can easily and accurately adjust the focal position, the inkjet head 12 can be easily and accurately positioned.

  Further, by using the bifocal microscope 34 that recognizes images of the positioning mark (reference nozzle 22) and the reference mark 42, the focus of each mark 22, 42 can be adjusted. A focused image can be obtained simultaneously with high accuracy, and alignment can be performed with high accuracy and ease.

  In the present embodiment, the proper movement position of the focus lens 86 is determined using the focus evaluation value for each movement amount of the focus lens 86. However, the focus evaluation value for each movement position of the focus lens 86 is used to determine the focus. The proper movement position of the lens 86 may be determined. Further, the proper movement position of the focus lens 86 may be determined using each movement amount of the focal position or a focus evaluation value for each movement position.

  Although omitted in the present embodiment, when focusing on the reference nozzle 22, the height of the nozzle surface NF may be adjusted as the focus position, or the middle portion of the reference nozzle 22 may be adjusted as the focus position. good.

  Furthermore, in the present embodiment, the present invention is applied to the bifocal microscope 34 described above, but the present invention is not limited to this as long as it is an image recognition optical system. For example, the present invention may be applied to other types of microscopes, and the present invention may be applied to various cameras.

  Furthermore, in the present embodiment, in the appropriate position determination operation, the reduced focus evaluation value is detected after all the focus evaluation value measurements are completed. However, every time one arbitrary focus evaluation value is measured, Then, it is determined whether or not the focus evaluation value is a reduced focus evaluation value. If it is determined that the focus evaluation value is a reduced focus evaluation value, the movement of the focus lens 86 is ended, and the measurement of the focus evaluation value is ended. The structure which transfers to the following process (extraction evaluation value extraction process: S3) may be sufficient. According to this configuration, the tact time of the focus evaluation value measurement process can be shortened.

  2: head alignment device, 12: inkjet head, 22: reference nozzle, 31: alignment jig, 33: moving stage, 34: bifocal microscope, 35: alignment mechanism, 36: alignment mask, 42: reference mark, 71: First optical system 72: Second optical system 74: Imaging unit 86: Focus lens 87: Lens moving mechanism NF: Nozzle surface

Claims (4)

A focus position adjustment method for an image recognition optical system that adjusts a focus position by moving a focus lens of an image recognition optical system in order to obtain a focused image of a recognition object,
A plurality of focus evaluation values that indicate the sharpness of the focused image with respect to the amount of movement of the focus lens while moving the focus position from the front of the assumed proper focus position to the front by the movement of the focus lens. A measurement process for measuring points;
A reduced evaluation value detection step for detecting a reduced focus evaluation value, which is a focus evaluation value that starts to decrease through a peak with the amount of movement of the focus lens, from among the plurality of focus evaluation values measured;
An effective evaluation value extracting step of extracting a plurality of focus evaluation values that are equal to or greater than the reduced focus evaluation value among the plurality of measured focus evaluation values as an effective evaluation value;
A focus position adjustment step of generating an approximate curve from the plurality of effective evaluation values, and moving the focus lens by the amount of movement of the peak to adjust the focus position. A method for adjusting the focal position of an optical system. An image recognition optical system capable of adjusting a focal position by moving a focus lens of an image recognition optical system in order to obtain a focused image of a recognition object,
A plurality of focus evaluation values that indicate the sharpness of the focused image with respect to the amount of movement of the focus lens while moving the focus position from the front of the assumed proper focus position to the front by the movement of the focus lens. Measuring means for measuring points;
A reduced evaluation value detecting means for detecting a reduced focus evaluation value, which is a focus evaluation value that starts to decrease through a peak with the amount of movement of the focus lens, from a plurality of measured focus evaluation values;
Effective evaluation value extraction means for extracting a plurality of focus evaluation values that are equal to or greater than the reduced focus evaluation value among the plurality of measured focus evaluation values as effective evaluation values;
An image recognition optical comprising: a focus position adjusting unit that generates an approximate curve from the plurality of effective evaluation values, moves the focus lens by an amount of movement of the peak, and adjusts the focus position. system. A head alignment device for positioning a plurality of inkjet heads having alignment marks on a nozzle surface,
An alignment mask having a plurality of reference marks corresponding to the alignment marks of the plurality of inkjet heads;
A mask fixing jig for positioning and fixing the alignment mask on a set table;
The image of the alignment marks of the plurality of inkjet heads and the reference marks of the alignment mask set in a pre-aligned state on the mask fixing jig is simultaneously recognized from below. Image recognition optical system,
A head alignment apparatus comprising: position correction means for correcting the position of each of the plurality of inkjet heads based on a recognition result of the image recognition optical system.   The image recognition optical system includes a bifocal microscope including a first optical system for recognizing the alignment mark and a second optical system for recognizing the reference mark. Item 4. The head alignment apparatus according to Item 3.

Images