JP4409467B2 - Eyeglass lens suction jig mounting device - Google Patents

Description

  The present invention relates to an eyeglass lens suction jig mounting apparatus that determines the mounting position of a suction jig to an eyeglass lens from an image of the eyeglass lens taken by an imaging means.

  For various raw spectacle lenses, the spectacle lens surface is imaged using an imaging device, image processing is performed, the presence or absence of figures is determined, and a suction jig such as a suction cup is attached according to the type of spectacle lens. Various techniques have been proposed for attaching an eyeglass lens suction jig of an eyeglass lens for automatic mounting.

  In particular, those with hidden marks written on fabric lenses such as progressive multifocal lenses, those with mark marks recorded on spectacle lenses, and those with segments (small balls) formed such as bifocal lenses are hidden. A device that picks up a graphic image of a reference mark using marks, print marks, and segments as a reference mark, and attaches a suction jig such as a suction cup to the suction position (optical center) of a spectacle lens that is geometrically related to the reference mark Is known (patent documents 1 to 6). The mark mark is a mark marked on a spectacle lens with a lens meter or the like, for example.

There is also known an apparatus for capturing a graphic image of a spectacle lens and binarizing an image signal to detect a concealed mark or the like imprinted or printed on the spectacle lens surface or the like, or irregularities or scratches on the spectacle lens surface ( Patent Documents 1-3, 5-7).
JP 2002-296144 A JP 2000-19058 A Utility Model Registration No. 3077054 DE3829488A1 publication JP 2002-139713 A JP 2002-1638 A EP856728A2 publication

  However, the spectacle lens may be engraved with a hidden mark that is difficult to see with the naked eye that displays position information unique to the lens. In some cases, the hidden marks are hidden by paint, and in any case, it has been troublesome to specify these positions.

  Then, it is necessary to check the presence or absence of the hidden mark and determine the attachment position of the suction jig based on the position of the hidden mark or the paint (fitting mark). In addition, some spectacle lenses such as progressive multifocal lenses have a fitting mark painted near the distance point or mounting position of the progressive multifocal lens, but if the presence or absence of a hidden mark cannot be confirmed, It is necessary to confirm the presence of the fitting mark and specify the suction position.

Therefore, the present invention can confirm the presence or absence of the hidden mark or paint of the progressive multifocal lens with higher accuracy, and if there is a hidden mark, the attachment position of the suction jig is automatically determined based on these positions. If the hidden mark or paint cannot be confirmed, check the presence or absence of the fitting mark painted near the distance point or the mounting position, identify the position of the mark, and perform the adsorption treatment. An object of the present invention is to provide a spectacle lens suction jig mounting device capable of automatically determining the mounting position of a tool .

In order to achieve this object, the invention according to claim 1, the mounting table on which the spectacle lens can be arranged in the opening, the imaging means for capturing an image of the spectacle lens mounted on the opening, In the spectacle lens suction jig mounting device, comprising: a position determining means for specifying the mounting position of the suction jig from the image of the spectacle lens; and a mounting means for arranging the suction jig at the mounting position of the spectacle lens. position determining means comprises a mark detecting means for detecting the presence and position of the horizontal reference hidden marks from the opening inside the image captured by the paint eyepoint become hidden marks and the spectacle lens of the spectacle lens in the imaging unit a mounting position determining means for determining the mounting position of the suction jig in the spectacle lens from the position the detecting, the hidden marks as a eyepoint of the spectacle lens Serial or as a reference for the mounting position of the suction jig, or characterized in that the horizontal reference hidden marks by painting the spectacle lens and a to switch the switching means as a reference for the mounting position of the suction jig.

According to a second aspect of the present invention, there is provided a mounting table that can be placed in the opening of the spectacle lens, an imaging unit that captures an image of the spectacle lens placed in the opening, and the captured image of the spectacle lens. In the spectacle lens suction jig mounting device, comprising: a position determining means for identifying the mounting position of the suction jig from the mounting means; and a mounting means for arranging the suction jig at the mounting position of the spectacle lens. the marked position detection means, said detection for detecting the presence and position of the horizontal reference hidden marks from the opening inside the image captured by the paint eyepoint become hidden marks and the spectacle lens of the spectacle lens in the imaging unit Mounting position determining means for determining the mounting position of the suction jig in the spectacle lens and the eyeglass lens eyepiece when detected by the mark detecting means. Screen display setting means for setting whether to display a screen based on a hidden mark as a reference or a screen display based on a horizontal reference hidden mark by painting of the spectacle lens. A tool mounting device is provided.

According to a third aspect of the present invention, in the apparatus for attaching a spectacle lens suction jig according to the first or second aspect, the mark detection unit performs binarization processing and noise removal processing on an image acquired by the imaging unit. And obtaining an edge image of a hidden mark to be an eye point on the spectacle lens or a horizontal reference hidden mark by painting of the spectacle lens .

Further, according to a fourth aspect of the present invention, in the eyeglass lens suction jig mounting device according to the third aspect, the mark detection means matches the edge image acquired by the imaging means with a mark template prepared in advance. To obtain a correlation between the edge image and the template and detect a portion having the highest correlation value as the hidden mark .

According to a fifth aspect of the present invention, in the spectacle lens suction jig mounting device according to the fourth aspect, the position determining means is an image of a hidden mark that becomes an acquired eye point or a horizontal reference hidden mark of the spectacle lens. The suction jig mounting location is specified from the position.

According to a sixth aspect of the present invention, in the apparatus for attaching a spectacle lens suction jig according to the fifth aspect, the position determining means detects a luminance value of a pixel of the image of the captured spectacle lens and detects the luminance value. When there is a pixel with the luminance value of 0, it is determined that the image includes a paint mark, and the coordinate average of the pixel portion with the luminance value of 0 is obtained or detected as the center of gravity of the paint mark. When there is a pixel having a predetermined luminance value, it is determined that the image includes a hidden mark, the coordinate average of the portion of the pixel having the predetermined luminance value is obtained as a centroid of the hidden mark, and the centroid The mounting location of the suction jig is specified from the above.

According to the present invention, a spectacle lens is imaged by the imaging unit graphic image, a hidden mark or paint imaged openings near graphical image spectacle lenses, in particular the position of the distance point of the progressive multifocal lens Can be detected with high accuracy, and a fitting mark near the distance point or the mounting position can be detected, and the mounting position can be specified based on the mark. In addition, it is possible to reduce errors in the mounting position due to manufacturing errors and deviations such as hidden marks and paints, etc., and to reduce mounting position errors due to fitting mark errors and deviations, etc. The position can be determined accurately and quickly .

The following embodiments of the suction jig mounting position determining method of the suction jig mounting apparatus and the spectacle lens of the spectacle lens of the present invention by referring to the drawings will be described in order below.
[Configuration of each part of the device]
FIG. 67 is a block diagram for explaining the outline of the configuration of the eyeglass lens suction jig attaching device 100 according to the present invention. In this example, the eyeglass lens suction jig attachment device 100 includes an entire optical system 110, a suction jig transport unit 120, a lens meter optical system (CL optical system) 130, a lens placement unit 140, and a lens placement unit 140. An XY stage unit 150 that moves back and forth, right and left, an operation panel 10, a liquid crystal display unit (display unit) 160, an arithmetic control circuit (arithmetic control unit) 170 as an image processing unit having an image processing function, and a ROM 170a are provided. .

  The ROM 170a controls the image processing program, the entire optical system 110, the suction jig conveyance unit 120, the lens meter optical system 130, the lens placement unit 140, the XY stage unit 150, the liquid crystal display unit 160, and the like. Is stored. The arithmetic control circuit 170 then executes an image processing program, the entire optical system 110, the suction jig transport unit 120, the lens meter optical system 130, the lens placement unit 140, and the XY stage unit according to the program stored in the ROM 170a. 150, the liquid crystal display unit 160, and the like are controlled.

<Operation panel 10>
The operation of the operation panel 10 enables display switching of the liquid crystal display 160 and various setting operations. That is, as shown in FIGS. 70 to 71, the operation panel 10 includes an operation panel unit 10 a disposed on the right side of the liquid crystal display 160 and an operation panel unit 10 b disposed on the lower side of the liquid crystal display 160. Have.
(Operation panel unit 10a)
The operation panel 10a includes a “stop” switch 12 for stopping the measurement, an “input switching / menu” switch 13 for switching the layout data input method, and a “memory” switch for calling frame data stored in the memory. 14, a “data request” switch 15 for requesting frame data, an “− +” switch 16 for input setting, and a “▽” switch 17 for cursor movement.

  The “input changeover / menu” switch 13 can be displayed for a predetermined time (several seconds, for example, 2 seconds) or longer to display a menu screen.

  Further, the “input switch / menu” switch 13 is used for instructing confirmation after manual positioning or position setting when the block 13 is waiting for a block instruction (suction instruction) and is pressed in a stop state after measurement. It can be done.

  When the “memory” switch 14 is pressed in the hidden mark observation mode, the screen of the liquid crystal display 160 is switched to the hidden mark storage screen.

  The “data request” switch 15 is used to request transfer of target lens shape data (θi, ρi) from a frame shape measuring device (not shown) connected to the lens suction jig mounting device 1.

  The “− +” switch 16 is used for setting the increase / decrease of numerical data in the portion displayed on the liquid crystal display 160 and whose display color is highlighted by the “▽” switch 17. The “− +” switch 16 is also used for switching the display magnification of the liquid crystal display 160 during manual alignment.

The “▽” switch 17 is used to move the cursor of the data input unit displayed on the liquid crystal display 160. The cursor here means that one of the display colors of the plurality of data input frames (data input portions) displayed on the liquid crystal display 160 is inverted or changed to another color. The state in which data can be input.
(Operation panel unit 10b)
The operation panel 10b is provided with function keys F1 to F6 arranged along the lower edge of the liquid crystal display 160. In addition, the operation panel unit 10b is provided with a “left” switch 18L and a “right” switch 18R for specifying the processing for the right eye and the left eye of the spectacle lens and switching the display.

<Function keys F1 to F6>
The function keys F1 to F6 are used at the time of setting related to the processing of the spectacle lens ML, and are used for responding to and selecting messages displayed on the liquid crystal display 160 in the processing step.

<Layout screen>
Further, as shown in FIG. 70, the layout screen includes a “layout / suction” mode for displaying a layout screen for sucking a lens suction jig on a spectacle lens, and target lens shape information (θi, ρi). There is a “layout” mode that shows the state when the lens suction jig is attracted to the hidden spectacle lens.

  When the “layout” tab TB1 is selected, the message display area E1, the numerical value display area E2, and the state display area E3 are displayed.

  In this example, the arithmetic control circuit 170 implements image processing by reading and executing an image processing program stored in the ROM 170a by the CPU. The eyeglass lens suction jig attaching device 100 is provided with a CF card input / output unit 171, an external connection interface unit 172, and a power supply device 173.

<Whole optical system>
The overall optical system 110 is a light-emitting diode 103 (for example, wavelength 860 nm) that is an optical system that captures a graphic image inside the spectacle lens L and the opening. A collimator lens 104 for light, two CCD cameras 105 and 106 as image sensors, two half mirrors 107 and 108 for guiding reflected light to the image sensor, and a reflector made of a retroreflective sheet rotated by a motor 111 109. In this example, the CCD camera 105 captures the entire lens, and the CCD camera 106, which is an image sensor, magnifies (for example, doubles) a part of the lens. Further, the reflector 109 is provided with an inclination of, for example, about 3 ° with respect to a vertical plane of the axis of the motor 111 in order to uniformly irradiate the lens.

  In this optical system, the light emitted from the light emitting diode 103 is converted into a parallel light beam via the collimator lens 104, and the reflected light from the reflection plate 109 is reflected on the spectacle lens by the CCD camera via the collimator lens 104 and the half mirrors 107 and 108. A graphic image is captured. At this time, unprocessed circular spectacle lenses and processed spectacle lenses have a mark mark attached by a lens meter, an engraved hidden mark or a printed print mark, and a bifocal lens has a semicircular segment ( (BF lens) and the like, the optical system is arranged so as to form an image near the spectacle lens surface.

<Lens placement part>
An opening 141 is provided at the center of the mounting stage. As shown in FIG. 68, an acrylic substrate 201 and three lens holding rods 202 for mounting spectacle lenses are provided on the substrate 201. A lens holder 200 is provided.

  In addition, as shown in FIGS. 17 and 67, the opening 141 is provided with three pins 142 (in FIG. 67, two pins due to the cross section) 142 that rotate in conjunction with each other. L or the side surface of the frame replacement lens holder 300 in FIG. 69 is sandwiched. Since these pins 2 are rotated in conjunction with each other, they are rotated by a single belt, and three of them are moved simultaneously.

  Since three pins (holding pins) (not shown) correspond to reference numeral 1302 in FIG. 17, for convenience of illustration, the portions indicated by reference numeral 1302 are assigned reference numerals for explanation. .

  Further, a claw portion is provided on a part of the ring-shaped flat plate covered on the three clamping members adjacent to the opening, and this claw portion cuts off the light across the optical switch portion. In this way, the origin of the device is calibrated.

<Frame replacement lens holder>
In the case of an unprocessed circular eyeglass lens, it can be placed by placing it on the above-mentioned holding member on the transparent plate, but a lens shape such as a processed eyeglass lens or a non-circular eyeglass lens is used as the holding member. The frame replacement lens holder 300 is used to place the lens on the screen. This is because, when the spectacle wearer wants to change the spectacle frame, the suction jig 180 is attached to the processed spectacle lens in order to frame the currently processed spectacle lens in a new spectacle frame.

  As shown in FIG. 69, the frame replacement lens holder 300 includes a lens holding arm 302 whose base end portion is pivotally supported on an annular substrate 301 so as to be rotatable about a shaft 303. The free end of the lens holding arm 302 is urged to rotate toward the center of the annular substrate 301 by a coil spring 302a.

  The annular substrate 301 is provided with a light transmitting hole 304 for recognizing the frame replacement lens holder 300. The frame replacement lens holder 300 is attached to the opening 141 in place of the lens holder 200 attached to the opening of the lens mounting table described above.

<XX stage part>
The XY stage unit 150 moves the lens mounting unit 140 in the X direction (advance / retreat direction) and the Y direction (lateral direction), so that the eyeglass lens L is taken out, the imaging position, and the eyeglass lens L are absorbed and cured. It is moved to the attachment position of the tool 121 and the lens meter optical system measurement position. At the imaging position, the position of the spectacle lens L imaged by the CCD cameras 105 and 106 is moved in the X and Y directions to change the imaging position.

<Suction jig transport section>
In FIG. 67, the suction jig transport unit 120 has a cylindrical cupped groove (not shown) and a suction cup transport unit (not shown) including a drive motor (not shown). This suction cup transport unit (not shown) is attached to a flat plate in the same manner as the entire optical system. Although detailed illustration of this suction jig conveyance unit 120 is omitted, it is configured as follows.

  That is, the transfer arm provided with a cup receiving portion for attaching a suction cup to the lower end is attached to the cylindrical portion, and the transfer arm descends downward by driving of the drive motor, and the suction attached to the cup receiving portion. The transfer arm is turned so that the suction surface of the cup faces downward and directly above the surface of the spectacle lens placed on the placement stage (mounting table).

  The suction part of the suction cup is constituted by the cylinder part, the arm part, the cup receiving part and the drive motor. The structure of the cup receiving portion is as shown in the figure. Conventionally, the base portion of the suction cup has only been sandwiched by wire-like springs built in the upper and lower portions of the cup receiving portion. When the suction jig is mounted, the suction cup is not easily detached from the cup receiving portion, and there is a problem that the spectacle lens is lifted through the suction cup. Therefore, an engagement plate portion that engages with a pin protruding from the cup receiving portion moving portion to the outer side surface is provided on the outer side surface of the cup receiving portion, and the lens transport arm descends downward to suck the suction cup on the spectacle lens surface. In addition, the pin is detached from the engagement plate portion through the pressing force applied to the cup receiving portion, and the suction cup is configured to be sucked and left by the spectacle lens. In addition, after the suction, the moving part of the cup receiving part moves, the light blocking of the optical switch provided at the rear end of the moving part is released, the light of the optical switch passes, the OFF switch is turned on, and the sucking work is performed. The operation processing circuit inside the apparatus determines that the process has been completed, the transfer arm returns to the upper rest position, and the lens mounting table returns to the initial position (a state in which the lens mounting table protrudes from the apparatus opening).

<Lens meter optical system>
The lens meter optical system 130 is a device for optically detecting the characteristics of spectacle lenses, and recognizes graphic images such as unprocessed circular unprocessed lenses, hidden marks, print marks, and BF lenses (segments). Used for eyeglass lenses that cannot. In this case, since the suction position of the suction jig cannot be specified, in the case of such a spectacle lens, the lens mounting table is moved in the Y-axis direction and moved to the CL optical system. As shown in the figure, the CL optical system includes a light emitting unit on the upper side and a light receiving unit on the lower side, and is configured to face each other vertically, and an optical system is disposed at the rear end of the light emitting unit.

[Action]
(I) Outline of Eyeglass Lens Image Processing Next, processing of the eyeglass lens graphic image by the arithmetic control circuit 170 of the eyeglass lens suction jig mounting apparatus 100 shown in FIG. 67 will be described. FIG. 1 is a flowchart showing the entire image processing of the eyeglass lens suction jig mounting apparatus according to the present invention.

  In this example, the arithmetic control circuit 170 acquires the whole image with the whole optical system 110 (S1), and determines whether or not there is a frame replacement lens holder 300 to which a non-circular lens is attached (S2). Next, the arithmetic control circuit 170 determines whether or not the spectacle lens L is present (S3), and further performs lens type determination (S4).

  Then, when the lens type is a progressive multifocal lens, the arithmetic control circuit 170 performs horizontal alignment of the lens in step S5, acquisition of an enlarged image in step S9, detection of a hidden mark (suction point / tilt) in step S10, and the like. Perform progressive multifocal lens processing.

  In addition, when the lens type is a lens with a segment (small ball) (bifocus lens), the arithmetic control circuit 170 performs segment detection (suction point / tilt) (S6), horizontal alignment of the lens (S6a), etc. ) Execute lens processing.

  Further, when the lens type is a lens with a marking point, the arithmetic control circuit 170 executes processing for the lens with a marking point, such as marking point detection (adsorption point / tilt) (S7) and horizontal alignment of the lens (S6a). .

  In addition, when the lens type is a markless lens, the arithmetic control circuit 170 drives and controls the XY stage 150 to move the spectacle lens L to the measurement unit of the lens meter optical system 130 to detect a suction point or the like. The non-marked lens is processed (S8).

  After each processing is completed, the spectacle lens L is moved to the attachment position (block standby position) of the suction jig 121 (S11), and there are dimensional relationships such as the lens radius, the frame shape, and the interpupillary distance of the spectacle wearer. It is determined whether it can be solved (radius data comparison process) (S12). Further, the left and right sides of the lens are confirmed (S13), a process of sucking the suction jig 121 (block suction execution process) (S14) is performed, and the series of processing ends.

(Ii) More Specific Description of (i) Image Processing by Arithmetic Control Circuit 170 Image acquisition based on the above-described flowchart of FIG. 1 will be described more specifically.
Step S1 (acquisition of whole image)

  By the way, in the above-mentioned step S1, when the eyeglass lens L on the lens holder 200 of the opening 141 and the opening 141 in FIG. 67 is entirely imaged by the CCD camera 105, as video data as shown in FIG. In gray scale, the image includes a shadow 1100 of the spectacle lens L, a shadow 1302 of the pin 142 that holds the spectacle lens L, a shadow 1201 of the lens holding rod 202, and a shadow 1500 of the distance center mark 1501 A shadow 1502 of the center mark for use and a hidden mark indicating the eye point are photographed.

In step S1, when the frame replacement lens holder 300 is used, the lens L, the shadow of the annular substrate 301 (not shown), and the shadow 1301 of the lens holding arm 302 (see FIGS. 7 to 12). Etc. are photographed.
Step S2 (determining whether there is a frame replacement lens holder)

  In the processing in step S2, the presence / absence of the frame replacement lens holder 300 used when the target lens is processed or non-circular is detected. That is, it is detected whether the frame changing lens holder 300 is attached to the opening 141 instead of the lens holder (lens holder) 200.

In this example, the presence / absence of the frame replacement lens holder is determined by detecting the light transmitting holes 304 provided around the frame replacement lens holder 300 shown in FIG.
Steps S21 to 23 (see FIGS. 2 and 3)

Specifically, the following processing is performed. First, when the frame replacement lens holder 300 is installed in the opening 141 in place of the lens holder (lens holder) 200, the position shown in FIG. A processing area Sa of 35 × 35 pix is set, the processing area Sa is cut out as shown in FIG. 3B (S21), the entire processing area is searched to obtain the maximum luminance value, and the value is set as the upper limit of the luminance value, With the value 0 as the lower limit, the contrast is increased as shown in FIG. 3C (S22), a threshold is set, and the processing region image is binarized to clarify the contrast as shown in FIG. Plan (S23). When there is an image of the light transmitting hole 304, an image 304a is obtained as shown in FIGS.
Steps S24 to S26 (see FIGS. 2 and 3)

Next, using a binarized template image of a hole prepared in advance as shown in FIG. 3E, a difference from the binarized processed image as shown in FIG. 3D is obtained, and the difference becomes zero. The number of pixels is counted (the number of pixels having the same value is counted) (S24). If the total number is equal to or greater than a predetermined threshold, it is determined that a hole is opened in the target processing area and a frame replacement lens holder is installed (S25, S26). If it is less than or equal to the threshold value, it is determined that the frame replacement lens holder 300 is not installed.
Step S3 (determining whether a lens is present)

  In this example, it is determined whether the spectacle lens L is placed on the stage. Although processing differs depending on the presence / absence of the frame replacement lens holder 300, basically, if there is something on the stage, it is determined that there is a lens.

If it is determined in step S2 that there is no frame replacement lens holder 300, it is not limited to the outer shape of the spectacle lens L, and whether there is something other than noise (paint mark, segment, hidden mark, lens, etc.) in the processing area. Judge with.
Steps S301 to S305 (see FIGS. 4 and 5)

  This determination is based on the procedure shown in the flowchart of FIG. First, as shown in FIG. 5, a processing area 1106 is cut out from the center of the stage to an edge in one direction that does not overlap the three pins on the stage (S301). Next, the pixels below the threshold value in these areas 1106 are binarized, the entire processing area is searched, and the number of pixels below the predetermined threshold value is counted (S302). If the counted number of pixels is equal to or greater than a predetermined value, it is determined that a lens is inserted (S303, S304). When the counted number of pixels is less than the threshold value, it is determined that there is no lens (S305).

  If it is determined in step S2 that the frame replacement lens holder 300 is present, a two-step determination is performed.

In this case, when the spectacle lens L is inserted and held in the frame changing lens holder 300, the lens holding arm 302 (indicated by reference numeral 1301 in the captured image) is considered to be open to some extent. Determines if there is an arm in the center of the stage. Next, a search is performed in six directions from the center of the stage, and if no edge is detected to the edge of the stage in at least one direction, it is determined that the lens is not inserted and held.
Steps S311-323

  The determination (judgment) processing for the presence or absence of the lens of the frame replacement lens holder 300 is performed according to the flowchart shown in FIG. First, an area 1107 at the center of the stage is designated and cut out (S311, FIG. 7), and the whole cut out processing area is searched. The number of pixels equal to or smaller than a predetermined threshold is counted (S312), and if the counted number of pixels is equal to or smaller than the threshold, the lens holding arm 302 does not exist in that region and the process proceeds to the next determination (S313, FIG. 8, FIG. 9). If it is greater than or equal to the threshold value, it is determined that there is an arm in the area and there is no lens (S320, S321), and the process ends.

Next, the lens outer shape is searched. That is, as shown in FIG. 10, the pixels are traced from the center of the stage toward both sides of the base of the shaft 303 of the lens holding arm 302 (S314). This search is performed for each of a total of six directions positioned toward both sides of the base of the shaft 303 of the three lens holding arms 302 as shown in FIG. In this process, the next pixel position is calculated from the current pixel (S315), the difference between the luminance at that point and the luminance at the previous position is calculated (S316), and the difference value is larger than a predetermined value and not near the stage. When (S317, S318), it is determined that there is a lens (S317, FIG. 11). This process is performed in all directions (S324), and when there is a large difference in the vicinity of the stage in at least one direction, it is determined that there is no lens (S322, S323, FIG. 12), and the process ends.
Steps S4 and 5 (Lens type judgment and lens horizontal alignment)

  Next, the determination of the lens type in steps S4 and S5 will be described more specifically based on the flowchart shown in FIG.

  First, an entire image is acquired as shown in FIG. 13 (S401). In the case of a progressive multifocal lens, the image shown in FIG. 17 is obtained, and in the case of a bifocal lens, an image including the segment LS shown in FIG. 15 is obtained.

  Next, in order to make the image clearer, the maximum value and the minimum value of the luminance value are set, and the luminance value of all the pixels is changed so as to obtain 256 gradations therebetween. In addition, a pixel having a luminance value equal to or higher than the maximum value is given as 255, and a pixel having a luminance value equal to or lower than the minimum value is given as a value. Thereby, the contrast is enhanced.

  Then, a LOG (Laplacian Of Gaussian) filter is applied to the entire image to extract lens outlines, paint marks, and segment outlines. The LOG filter performs smoothing by multiplying the luminance values of the target pixel and its neighboring pixels by a weight determined by a Gaussian function, and further applies a Laplacian to extract an edge.

  Further, for each pixel value obtained by the filtering process, a value equal to or higher than the given threshold value is converted into a luminance value 255, and a value equal to or lower than the threshold value is converted into a luminance value 0 to perform binarization (S402). Thereby, for example, a binary image as shown in FIG. 18 is obtained. In this example, the progressive multifocal lens is binarized.

  Many noises other than the necessary lens outline and paint mark are extracted from the image created in step S402. In order to remove it, labeling is performed to eliminate minute noise (S403). Here, labeling is a known process in which the same label number is assigned to the connected pixels for distinction.

  Next, label numbers with a small total number are removed as noise. At this time, the threshold value must be given so that the mark point is not regarded as noise. For example, an image shown in FIG. 19 is obtained by the noise removal process.

  Next, the geometric center of the spectacle lens L in FIGS. 67 and 68 is determined. As shown in FIG. 20, the lower left, lower right, and upper three points P1 to P3 are selected from the shadow 1100 of the outer shape of the spectacle lens L, the centers of the circles passing through these three points P1 to P3 are obtained, and the positions are determined as the temporary lenses. The center is O. This is for determining a search area such as line detection, and need not be exact. As shown by arrows A1 to A3 in FIG. 20, from the outermost part of the shadow 141a of the opening 141, the lower left and the lower right are searched in the horizontal direction, and the upper is in the vertical (down) direction. Suppose that the first edge is the edge of the lens.

  Next, a region (300 × 60 pixels: region shown in FIG. 21) centered on the temporary center (temporary geometric center) O of the lens obtained by the above processing is cut out as a horizontal paint line extraction processing region PA. In the image extracted by horizontal paint mark extraction (tilt calculation) in the region shown in FIG. 22, a straight line having the largest number of pixels with the luminance value 255 is obtained (Hough transform), and the straight paint is applied to the horizontal paint 1500 (four lines). Horizontal paint 1500a-1500d). However, if the number of pixels of the luminance value 255 is small on the extracted straight line, it is assumed that it is not horizontal painting because there is a possibility that it is a mark. Further, the inclinations of the horizontal paints 1500a to 1500d obtained here are obtained. The lenses that have confirmed the horizontal paints 1500a to 1500d are determined to be progressive multifocal lenses (S406). Next, a hidden mark is detected for the lens determined to be a progressive multifocal lens by the above processing. In this process, the position of the hidden mark is detected and whether or not the hidden mark position is painted is determined (steps S5 to S10). In this process, a hidden mark is detected and an adsorption point is determined.

  First, before the enlarged image is acquired, the inclination of the lens is roughly corrected (S5). Using the temporary geometric center of the lens and the tilt of the lens obtained in the lens type determination (S4), the stage is moved and rotated so that the horizontal paint mark of the lens is horizontal and the temporary geometric center is at the center of the image. Is. Next, an enlarged image is acquired. At this time, a double magnified image is acquired by the CCD camera 106 for magnified photographing so that the hidden mark can be seen. The obtained image is shown in FIG. This image is acquired in the same gray scale as the above-described image. The image includes a shadow 1201 of the lens holding rod 202, four horizontal paints 1500a to 1500d, a shadow 1501 of the distance center mark 1501, a shadow 1502 of the near center mark, between the horizontal paints 1500a and 1500b, and the horizontal paint 1500c, Two horizontal reference hidden mark shadows 1503 and 1503 positioned between 1500d, a center shadow 1504 positioned between the two horizontal paints 1500b and 1500c, and a hidden mark shadow 1505 indicating an eye point appear.

  Next, a hidden mark extraction area is set. From the standard of a general spectacle lens L, it can be estimated that the horizontal reference hidden mark is in the vicinity of a point 17 mm (about 170 pixels) in the horizontal direction from the center of the image. Therefore, as shown in FIG. 24, a rectangular frame 1600 of 50 to 70 pixels long × 90 to 110 pixels wide centering on the point is extracted as a hidden mark search region.

    When the shadow 1505 (fitting mark) of the hidden mark indicating the eye point is extracted, instead of the hidden mark area extraction frame 1600 of FIGS. 24 and 70, the distance center mark as shown in FIG. 1501 and a region extraction frame 1600a that surrounds the vicinity of the mark point position and a very small fitting frame 1600b positioned at or near the lower end of the distance center mark 1501 are switched so as to be superimposed on the image of the eyeglass lens. You can also.

  Furthermore, when the hidden mark shadow 1505 (fitting mark) indicating the eye point is extracted, the screen switching is set so as to switch to the screen displaying the hidden mark area extraction frame 1600 of FIG. 71, the area extraction frame 1600a that surrounds the vicinity of the distance center mark 1501 and the mark point position and the very small fitting frame 1600b that is positioned at or near the lower end of the distance center mark 1501 are displayed. Initial screen setting (learning function setting) can also be performed so as to switch to the screen.

  In this case, when the function key F6 corresponding to “Return” is pressed, the arithmetic control circuit 170 causes the liquid crystal display 160 to display the menu screen of FIG. 71A as the initial screen setting (learning function setting). In this menu screen, “Setting machine state setting and select item” etc. are displayed at the top, and menus Ma and Mb are displayed at the bottom corresponding to function keys F1 and F3. In addition, “execute” and “return” are displayed corresponding to the function keys F5 and F6.

  In the menu Ma, “optical center input method, minimum required lens diameter display, cup detection, cup size determination, processing condition setting input, measurement image left / right comparison, start switching, mirror inversion designation, progressive Selection items such as “detect lens” are displayed by the arithmetic control circuit 170.

  In addition, on the menu Ma, the arithmetic control circuit 170 displays a point cursor P1 with the background color of the selection item reversed or other colors. The selection item displayed on the menu Ma can be selected by pressing the function key F1. That is, in the menu Ma, every time the function key F1 is pressed, the point cursor P1 moves the optical center input method, the minimum required lens diameter display, the detection of the cup, the determination of the size of the cup, the input of the processing condition setting, and the comparison of the measurement image on the left and right , Start switching, mirror inversion designation, and progressive lens detection. The position where the point cursor P1 is located becomes the selected item.

  Similarly, in the menu Mb, the arithmetic control circuit 170 displays a point cursor P2 in which the background color of the selected item is reversed or other colors. When “optical center input method” is selected in the menu Ma, items such as “input” and “not input” are displayed in the menu Mb, and “progressive lens detection” is selected in the menu Ma. Then, items such as “hidden mark” and “paint mark” are displayed on the menu Mb.

  Moreover, selection of the selection item displayed on the menu Mb can be performed by pressing the function key F3. That is, while the items such as “input” and “not input” are displayed on the menu Mb, the point cursor P2 alternately moves between “input” and “not input” each time the function key F3 is pressed. Then, one of “input” and “not input” is selected. In addition, while the item “hidden mark”, “paint mark”, etc. is displayed on the menu Mb, the point cursor P2 moves alternately to “hidden mark” and “paint mark” each time the function key F3 is pressed. Then, one of “hidden mark” and “paint mark” is selected.

  After the selection of the menu items Ma and Mb is completed by such an operation, the setting is executed by pressing the function key F5, and the screen display of the liquid crystal display 160 is set.

  Once the screen switching is set in this way, the next time the screen is switched, the learning function is exhibited so as to switch to the set screen, and the screen for displaying the set hidden mark area extraction frame 1600 of FIG. 70 or the screen of FIG. It is possible to switch to a screen that displays the distance center mark 1501 or the area extraction frame 1600a surrounding the vicinity of the mark point position and the very small fitting frame 1600b located at or near the lower end of the distance center mark 1501. it can.

  In this way, the arithmetic control circuit 170 and the function keys F1, F3, F5, F6 and the like function as screen display setting means.

  In FIGS. 70 and 71, Ls indicates a lens shape based on lens shape (lens shape) information such as a spectacle frame, a demo lens, or a template, and Rp indicates a lens suction jig displayed superimposed on the lens shape Ls. The jig shape is shown. Further, in the vicinity of the mark point position, a linear paint 1600c extending in the horizontal direction so as to contact the lower end of the distance center mark 1501, and a shadow of a hidden mark indicating an eye point serving as a fitting mark (not shown in FIG. 71). The shadows 1505) shown in FIGS. 24 and 25 are painted. This shadow (shadow 1505 shown in FIGS. 24 and 25) is located in the fitting frame 1600b in the region extraction frame 1600a.

  With this fitting frame 1600b, it is possible to clearly show the operator where the shadow (the shadow 1505 shown in FIGS. 24 and 25) that is the fitting mark is.

  Further, on the basis of the fitting frame 1600b, a shadow of a hidden mark indicating an eye point serving as a fitting mark (shadow 1505 shown in FIGS. 24 and 25, not shown in FIG. 71) is searched, and the eye serving as the fitting mark is searched. The suction center data at a position where the center of the jig image Rp of the suction jig 121 of FIG. 67 coincides with the shadow of the hidden mark indicating the point (shadow 1505 shown in FIGS. 24 and 25 omitted in FIG. 71). It is determined by the control circuit 170. Then, the arithmetic control circuit 170 displays the jig image Rp so as to overlap the lens shape Rs based on the suction center data obtained by this calculation.

  Therefore, the arithmetic control circuit 170 drives and controls the XY stage unit 150 and the suction jig transport unit 120 in FIG. 67 based on the obtained suction center data, and the center of the suction jig 121 in FIG. 67 is controlled. The suction jig 121 can be attached to the eyeglass lens L at a position corresponding to the suction center data of the eyeglass lens L.

 Incidentally, switching from FIG. 70 to FIG. 71 can be switched to the screen shown in FIG. 71 by pressing the function key F1 corresponding to “switch” in the lower left of the screen of FIG. In addition, in FIG. 71, it is possible to return to the screen of FIG. 70 by pressing the function key F6 corresponding to “Return” at the lower right of the screen.

  If the shadow of the hidden mark indicating the eye point to be the fitting mark (shadow 1505 shown in FIGS. 24 and 25, not shown in FIG. 71) is not extracted, the interval between the two horizontal paints 1500b and 1500c is not obtained. The shadow of the hidden mark indicating the eye point serving as the fitting mark with reference to the shadow 1504 of the center located at the center 1502 and the shadows 1503 and 1503 of the two horizontal reference hidden marks (in FIG. 24 and FIG. 25 not shown in FIG. 71) The shadow 1505) to be shown is specified, and the suction jig 121 is attached to the spectacle lens L so that the center of the suction jig 121 of FIG.

  First, binarization processing of image data is performed. The Canny operator is applied to the designated area to extract the edge, and the value is set to 255 when the value of each pixel is equal to or greater than a predetermined threshold value, and is set to 0 when the value is equal to or less than the threshold value. As a result, the value becomes 255 at the point where the change of the luminance value is abrupt. Next, after removing edges such as horizontal paint marks on the outer periphery of the region from the created binarized image and performing labeling and removing noise with a small number of connected pixels (FIGS. 25A and 25B). (FIG. 25 (c)), the shadow 1503 of the horizontal reference hidden mark is detected. This is done by creating a frame of about 27 × 27 pixels and performing a search in the area, and searching for a range having the largest edge amount in the processing area.

  Next, it is determined whether the detected mark is a horizontal reference hidden mark (horizontal paint 1500a to 1500d or shadow 1503 or the like) or another paint mark. That is, it is confirmed whether a paint mark is included at the position of the detected hidden mark. As shown in FIG. 26, binarization is performed by setting a low threshold (brightness value of about 100) for the range of the frame. As a result, if there is a pixel having a luminance value of 0 in the area, the center of gravity is calculated with the paint mark assuming that paint is included. FIG. 26A shows an example in which this process is performed and the horizontal reference hidden marks (horizontal paints 1500a to 1500d) are found, and FIG. 26B is different from the horizontal reference hidden mark by performing this process. It is an example that has been found to be another paint VG.

  Next, eye points are extracted. The coordinate average of the pixels having the luminance value 255 within the range obtained by the above processing is obtained and used as the center of gravity of the hidden mark. FIG. 27A shows an example in which the center of gravity G of the hidden mark is obtained. The coordinate average of pixels with luminance 255 is obtained from the binarized image obtained in the same manner as this processing, and the point is set as the center of gravity G of the paint mark VG as shown in FIG.

  Further, the inclination of the lens is obtained. The inclination and center position of the lens are obtained from the coordinates of the obtained two hidden marks. With a cross mark template as shown in FIG. 28, as shown in FIG. 29, matching is performed at points 0 mm, 2 mm, and 4 mm above the horizontal position in the vertical direction of the horizontal paint mark. It is assumed that the position with the highest correlation value is the position with the cross mark indicating the eye point. If all the correlation values are below a certain value, it is assumed that there is no cross mark indicating an eye point. The processing in that case will be determined separately.

  Next, the segment of the bifocus lens is detected, and the attachment position of the suction jig 121 is determined. First, an edge image of a segment is acquired. Here, the center of gravity and the number of connections for each label are calculated during the labeling.

  When the image shown in FIG. 15 is used, the spectacle lens L has a label 1: labels 2, 3, and 5 of the lens L corresponding to the shadow 1201 of the lens holding rod 300 in FIGS. 1401 areas are allocated.

  The label 1 in FIG. 15 shows the entire spectacle lens L. However, when the label 1 in FIG. 15 is the center of gravity at the center of the spectacle lens L, in FIG. It is displayed at the indicated point. Further, the center of gravity of the position indicated by the labels 2, 3 and 5 of the lens L is represented by a point surrounded by a circle 1411 in FIG. Furthermore, the center of gravity of the region of the segment edge 1401 is represented by a point surrounded by a circle 1412 in FIG.

  These barycentric coordinates, number of connections, and size are shown in Table 1.

次に、ラベルの重心が入力画像のある範囲内にあるかどうかを確認する（Ｓ４１３）。この処理は、図１６に示すように、枠１４１３内に重心があるものをセグメントの候補とする。セグメントの重心は必ず画像全体の下半分の中央付近にあると予想できるから行なる。さらに、ラベルの連結数がある範囲内にあるかどうかを確認する（Ｓ４１４）。予め、セグメントの連結数はだいたい４５０〜２０００画素程度であることを確認しているからこの範囲にあるものをセグメントとして抽出する。さらに、それぞれのラベルで右の様に幅と高さを測る（Ｓ４１５）。これは、図３３に示すようにセグメントのエッジ（輪郭線）１４０１に接する水平線Ｌａ１，Ｌａ２及び垂直線Ｌｂ１，Ｌｂ２を引きその間隔をそれぞれ高さ及び幅として検出する。この幅及び高さは、レンズが傾いていたとしても１
００×１００〜３００×３００画素程度の範囲に納まることを確認している。

Next, it is confirmed whether the center of gravity of the label is within a certain range of the input image (S413). In this process, as shown in FIG. 16, a candidate having a center of gravity within a frame 1413 is set as a segment candidate. This is because the center of gravity of the segment can always be expected to be near the center of the lower half of the entire image. Further, it is confirmed whether or not the number of connected labels is within a certain range (S414). Since it is confirmed in advance that the number of connected segments is about 450 to 2000 pixels, the segment is extracted as a segment. Further, the width and height are measured as shown on the right for each label (S415). As shown in FIG. 33, horizontal lines La1 and La2 and vertical lines Lb1 and Lb2 that are in contact with the edge (contour line) 1401 of the segment are drawn and the intervals are detected as height and width, respectively. This width and height is 1 even if the lens is tilted.
It has been confirmed that it falls within the range of about 00 × 100 to 300 × 300 pixels.

  If all the determinations (S413 to S415) are satisfied, the edge with the label is regarded as the edge of the segment, and it is determined that the lens is a bifocal lens (S416). If all the labels are checked and there is nothing that satisfies all the determinations, it is determined that the lens is not a bifocal lens, and the determination is made for another lens.

  As a determination of other lenses, a lens with a mark is determined (S407). This processing is performed when neither a progressive multifocal lens nor a bifocal lens is used.

  In the process of the lens with a dot, first, a labeled binarized image is acquired, and in the processing at the time of labeling (step S403), each center of gravity is obtained. If there is a label with the center of gravity at the midpoint, these are regarded as the three mark points. Then, all labels are checked, and if there is no mark, it is determined as a markless lens.

  Next, segment detection of the bifocal lens is performed (step S6). In this process, the segment of the bifocal lens is detected to calculate the suction point coordinates and inclination. This process is performed according to the flowchart shown in FIG.

  This will be described in detail below. Here, as shown in FIG. 32, if the segment connecting two corner points 1402 and 1404 in the segment 1401 is Lc1, and the vertical bisector of the segment Lc1 is 1406, the suction position 1407 is the vertical bisector. The intersection point of the upper curve of the line 1406 and the edge 1401 of the segment. Further, as shown in FIG. 31, the inclination θ of the lens is the inclination θ of the line segment connecting the two corner points 1402 and 1404. Next, the center of gravity is obtained from the edge of the segment, the distance from the center of gravity to the edge is taken, and a distance function is created. Then, the maximum value of the distance function is set as a corner.

  In this procedure, as shown in FIG. 30, an edge 1401 is first detected (S601, S631). This processing has already been described in the lens type determination (S4). The presence / absence of a segment is determined based on the detected edge 1401 (S602). Since the segment is located slightly below the center of the lens, the size is within a certain range, and the edge image is already labeled, using these, the center of gravity is determined for each labeled edge. The number of connections and the size are obtained, compared with a predetermined threshold value, and an edge satisfying these three conditions is regarded as the edge 1401 of the segment. Each threshold was set as follows by experiment.

Center of gravity: Range of 100 × 100 to 300 × 300 pixels near the center of the lower half of the lens image Number of connections: 450 to 2000 pixels (number of pixels for each label)
Size: 50 to 250 pixels for both height and width Note that, for the size, as shown in FIG. 33, the number of pixels of the height and width of the edge 1401 of the segment is measured.

  Next, the edge is thinned (S603). Before finding the center of gravity of the segment edge, the edge is thinned. The method of obtaining the center of gravity will be described in step S604. However, if there is a bias in the edge appearance (edge width unevenness), the position of the center of gravity may be shifted. Turn into. In this example, processing was performed in the vicinity of 8.

  Next, the center of gravity of the segment is detected (S604). From the thinned edge of the segment, the center of gravity G of the segment was obtained using the following equation.

ただし、ｘｉ,ｙｉはエッジ上のｉ番目（ｉ=０〜ｎ−１）の座標を示す。

Here, xi and yi indicate i-th (i = 0 to n−1) coordinates on the edge.

  Further, a distance function is created (S605). The distance Ln to each pixel on the edge of the segment is obtained from the center of gravity G obtained in step S604, and one round (360 °) from one predetermined point of the segment, for example, one corner point, is arranged as a distance function.

  Next, a corner portion is detected from the distance function (S606). Two maximum values are detected from the distance function obtained in step S605, and are detected as corner portions 1402 and 1404 therefrom.

FIG. 36 shows an image related to the sample of the bifocal lens in which corner detection has been performed. Symbols A to D in the figure indicate correspondences with symbols on the distance function. At this time, the center of gravity used for the calculation is indicated by a point G. Further, the obtained results are shown by circles in FIG.
Here, curve fitting is performed on the distance function to improve accuracy. The distance function is divided into two left and right regions at point C in FIG. 36, a quadratic function is fitted to each distance function, and an edge corresponding to the intersection is extracted as a corner.

  If the difference between the maximum value and the minimum value of the distance function is within a certain range, it is determined as a circle. As described above, the segment may be circular in addition to the half-moon shape that has been dealt with in experiments so far (see FIG. 39). The distance function at that time is shown in FIG. In this case, the distance function from the center of gravity differs from that of the half-moon shaped segment and is almost constant. Thus, when there is almost no difference between the maximum value and the minimum value of the distance function, it is determined that the segment is a circular segment, and corner detection is not performed.

  Next, detection of a corner when the edge of the segment LS of the spectacle lens L and the lens holding rod 202 overlap as shown in FIG. 41 will be described. In using this example, the lens holding rod 202 and the segment LS are inevitably overlapped with each other in order to correspond to lenses of various manufacturers. If the images are taken with these overlapping, an image shown in FIG. 42 is obtained. If an edge is detected from this image, the edge 1401 of the segment including the shadow 1201 of the lens holding rod 202 is detected, which affects the creation of a distance function from the center of gravity. Therefore, even in such a case, processing is performed so that the corner can be detected (estimated if it has disappeared). The general procedure is shown in the flowchart shown in FIG.

  In this example, the edge of the segment LS is detected (S621), and when the edge does not have an end point, the above-described normal processing shown in FIG. 30 is performed (S633). Process. First, correction for extending the edge is performed to make the edge a closed curve (S623). Next, a temporary center of gravity of the edge that becomes the closed curve is obtained, and a distance function is created (S624, S625). Then, when the edge shape is a circle, the processing when the segment is circular (S626, S634) is performed, and when the edge is not circular, the following processing is performed.

  If the edge is not circular, the number of remaining corners is calculated from the distance function. If there is only one remaining corner, an approximate edge is obtained by fitting a curve and an ellipse to the edge of the ball (S628), and the intersection of both lines is set as a corner (S629).

  If there are two remaining corners, the position is obtained (S630), and the corner position is calculated (S632).

  Next, how to count the remaining corners will be described.

Which part of the segment edge the lens holding rod 202 overlaps with is causing an edge break,
(A) When a part other than the corner is missing (FIG. 57: two corners remain),
(B) Let us consider two cases: the case where one of the left and right corners is missing (FIG. 59: one corner remaining). The distance function in each case is shown in FIGS. The number of inflection points, which is the maximum value of the distance function, is obtained and the number is regarded as the number of corners. However, as shown in FIGS. 58 and 62, since the distance function itself is very uneven, a weighted average of the distance function is taken here, and the distance function is obtained by using an inclination for each section.

  The distance function and the inflection point of the image shown in FIGS. 62 and 64 were obtained by the above method. The results are shown in FIGS. In each figure, the part indicating the inflection point which is the maximum value is indicated by a circle. As a result, it can be confirmed that there are two maximum values in the distance function of FIG. 63 and one maximum value in the distance function of FIG. Although not shown here, it was confirmed that the number of local maximum values can be obtained by this method also in other images.

Next, the detection of the remaining corners and the estimation of the position when hidden are described.
Since the position of the lens holding bar 202 is known in advance in the image, a mask is put on the position of the lens holding bar 202 to detect the edge. In addition, since it has been confirmed from the size of the three lens holding rods 202 and the segments that only one lens holding rod 202 can be hooked at most, here, the above-mentioned (A) and (B) are more specific. Think about it. That is, here, the above-mentioned (a) and (b) are more specifically
(A) When the lens holding bar 202 overlaps with a portion other than the corner,
(B) When it overlaps the lens holding rod 202 at the corner,
The distance function was determined in two cases.

Hereinafter, (a) and (b) will be described in more detail.
(I). In the case where the lens holding bar 202 overlaps at a portion other than the corner In this case, the image shown in FIG. 45 was processed. A distance function created from the obtained center of gravity G is shown in FIG. From this result, it was confirmed that even when the edge was interrupted by the lens holding rod 202, the maximum value was taken near the corner. However, since the edge of the portion overlapping with the lens holding rod 202 has disappeared, the center of gravity G obtained from the edge is close to the point C in FIG. Therefore, as compared with the distance function (see FIG. 37) when there is no overlap with the lens holding rod 202, it can be confirmed that there is a deviation in the magnitudes of the two maximum values (B and C) indicating the vicinity of the corner.
As a result, even when the edge is interrupted, the vicinity of the corner can be extracted by fitting the distance function to a quadratic function. However, in the distance function here, since the position of the center of gravity is shifted, the intersection of the fitted quadratic function may deviate from the original corner position. Therefore, a straight line is extended from each of points A and D in FIG. 44, a pseudo edge of the segment is created, a temporary center of gravity G ′ is obtained, and a distance function is calculated therefrom. The straight line extending from the end points A and D is determined by the least square method based on the coordinates of five points from the end points. The result is shown in FIG. 46, and the obtained distance function is shown in FIG. In the figure, the line segment AE and the line segment ED are added straight lines, and the symbol G ′ is a temporary center of gravity obtained from the pseudo edge.

The result of corner detection from the pseudo edge created here is shown as a circle in FIG. 48, and the result of fitting a quadratic function to the distance function is shown in FIG.
(B). When overlapped with the lens holding rod 202 at the corner The schematic processing procedure is shown in the flowchart of FIG. That is, in this process, two end points are detected from the image edge (S641), and a distance function is created (S642). Next, the remaining corner is detected (S643). At this time, the corner where the maximum value of the distance function remains is determined. Next, the edges are separated into two upper and lower edges (S644), and the approximate curve and ellipse of each edge are calculated, and the intersection of both is calculated as a corner (S646).

  Moreover, these processes were performed using the image shown in FIG. However, this image is obtained by adding a shadow of the lens holding bar 202 to a segment not overlapping the lens holding bar 202 in order to compare detection positions in later accuracy verification. FIG. 51 shows the edge detection result, and FIG. 53 shows the distance function. First, what kind of result will be obtained if a distance function from the center of gravity is created when the lens holding rod 202 overlaps the edge of a segment.

  From the obtained distance function (FIG. 53), it can be seen that the position of the remaining corner (point B) can be detected by taking the maximum value. However, since the other corner is hidden by the lens holding rod 202, an edge cannot be obtained and the corner cannot be detected. Therefore, here, as shown in FIG. 54, the corner position hidden by the lens holding rod 202 is estimated. That is, paying attention to the shape of the segment, the edges are separated into upper and lower curves at the corner, and an approximate curve or approximate ellipse is applied respectively, and the intersection is set as the corner.

  First, the edges are separated (S641). Based on the distance function shown in FIG. 53, the edge data is separated at the rough position (point B) of the remaining corner. After the edge separation, an approximate curve or ellipse formula for each edge is determined. The results of these approximations are shown in FIG.

  Here, since the image obtained by this method is an image obtained by adding the lens holding rod 202 later, in order to confirm the position where the result appears on the original image, the result is digitized to the original image. .

  FIG. 63 shows the obtained corner position. From this result, it was confirmed that even when the corner disappeared by overlapping the lens holding rod 202, an approximate curve or an ellipse was obtained and the corner position was estimated from the intersection.

In the above, the correspondence when the edge of the segment LS is interrupted by the lens holding rod 202 has been described. First, using the “count the number of remaining corners from the distance function” algorithm, it is possible to confirm which part of the segment the lens holding rod 202 overlaps to cause a break in the edge. Depending on the number of remaining corners, the position of the corner is detected from the maximum value of the distance function if it is 2, and the edge of the segment is approximated to a curve or an ellipse if it is 1, and the intersection of both lines is detected as a corner. Thus, the suction points of all kinds of spectacle lenses L are found. Next, the XY stage 141 is moved, and the eyeglass lens is moved to a place where the suction jig transporting part 120 of the suction jig installs the suction jig 121 (step S12). Further, the left / right determination of the lens (step S13) is performed.
(A) Progressive multifocal lens After the lens suction point is determined, whether the right and left selection of the lens performed by the user at the start is correctly performed is confirmed by image processing. The left / right determination of the lens is determined by whether the center of gravity of the near-end paint mark of the progressive multifocal lens is on the left or right side of the hidden mark midpoint.

In this process, first, as shown in FIG. 65, an area to be processed is cut out based on the midpoint of two hidden marks derived when the hidden mark is detected. Next, a search is made in the cut processing area, and a pixel value equal to or greater than a preset threshold value is determined as paint, and a coordinate value of the pixel is obtained. Furthermore, the average value of the obtained coordinates, that is, the center of gravity is obtained, and the right and left of the lens are determined depending on which of the hidden points is at the midpoint of the hidden mark.
(B) In the case of a bifocal lens This process takes a horizontal line and a vertical line from the shape of a segment photographed as shown in FIG. 66, compares the geometric center of the lens outer shape and the suction point, and determines which direction it is in. , Make a right / left decision. It was determined by the positional relationship between the geometric center O of the lens outer shape and the suction point 1407. If the geometric center is to the left of the suction point, it is determined to be the left lens, and if it is to the right, it is determined to be the right lens. After the above processing is completed, the suction jig transport unit 120 places the suction jig 121 on the suction point 1407 of the spectacle lens L, and the spectacle lens L is carried out of the apparatus by the suction jig transport unit 120. finish.

  As described above, the spectacle lens suction jig mounting device according to the embodiment of the present invention includes a mounting table (the lens mounting unit 140 or the lens holder 200) on which the spectacle lens L can be disposed in the opening 141, and Image pickup means (CCD cameras 105 and 106) for taking an image of the eyeglass lens L placed in the opening 141, and position determination for specifying the attachment position of the suction jig 121 from the taken image of the eyeglass lens L And means for mounting the suction jig 121 at the mounting position of the eyeglass lens L (including the XY stage unit 150 and the suction jig transport unit 120).

  In addition, the position determination unit detects the presence and position of at least one of a fitting mark or a hidden mark of the spectacle lens L from an image inside the opening 141 imaged by the imaging unit (CCD cameras 105 and 106). Means (arithmetic control circuit 170), mounting position determining means for obtaining the mounting position of the suction jig 121 on the spectacle lens L from the detected position (computing control circuit 170, including the XY stage unit 150), the spectacles There is provided switching means (function key F1) for switching between the fitting mark of the lens L (shadow 1505 indicating the eye point) or the hidden mark (1503) as a reference.

  According to such a configuration, the spectacle lens L is imaged by the image pickup means (CCD cameras 105 and 106), and the graphic image spectacle lens around the imaged opening 141 is hidden with a hidden mark or paint, and particularly progressive. The position of the distance point of the focusing lens can be detected with high accuracy, and the fitting mark in the vicinity of the distance point and the mounting position can be detected, and the mounting position can be specified based on the mark.

  In addition, it is possible to reduce errors in the mounting position due to manufacturing errors and deviations such as hidden marks and paints, etc., and to reduce mounting position errors due to fitting mark errors and deviations, etc. The position can be determined accurately and quickly.

  In addition, the spectacle lens suction jig mounting device according to the embodiment of the present invention includes a mounting table (the lens mounting unit 140 or the lens holder 200) on which the spectacle lens L can be disposed in the opening 141, and the opening 141. Imaging means (CCD cameras 105 and 106) for taking an image of the spectacle lens L placed on the camera, position determining means for specifying the attachment position of the suction jig 121 from the captured image of the eyeglass lens L, and suction Mounting means (including an XY stage unit 150 and a suction jig transporting unit 120) for arranging the jig 121 at the mounting position of the spectacle lens L is provided. In addition, the position determination unit detects the presence and position of at least one of a fitting mark or a hidden mark of the spectacle lens L from an image inside the opening 141 imaged by the imaging unit (CCD cameras 105 and 106). Means (arithmetic control circuit 170) and mounting position determining means for obtaining the mounting position of the suction jig 121 in the spectacle lens L from the detected position (comprising the arithmetic control circuit 170 and the XY stage unit 150). . Further, the position determining means may display the screen based on the fitting mark (hidden mark shadow 1505) of the spectacle lens L when detected by the mark detecting means (arithmetic control circuit 170), or Screen display setting means (arithmetic control circuit 170 and function keys F1, F3, F5, F6, etc.) for setting whether to display the hidden mark as a reference is provided.

  Once the screen switching is set in this way, the next time the screen is switched, the learning function is exhibited so as to switch to the set screen, and the screen for displaying the set hidden mark area extraction frame 1600 of FIG. 70, or FIG. The screen can be switched to a screen for displaying the distance center mark 1501 and the region extraction frame 1600a surrounding the vicinity of the mark point position and the very small fitting frame 1600b located at or near the lower end of the distance center mark 1501. In addition, according to such a configuration, the spectacle lens L is imaged by the imaging means (CCD cameras 105 and 106), and the graphic image spectacle lens around the imaged opening 141 is hidden with a hidden mark or paint. The position of the distance point of the progressive multifocal lens can be detected with high accuracy, and the fitting mark in the vicinity of the distance point and the mounting position can be detected, and the mounting position can be specified based on the mark. In addition, it is possible to reduce errors in the mounting position due to manufacturing errors and deviations such as hidden marks and paints, etc., and to reduce mounting position errors due to fitting mark errors and deviations, etc. The position can be determined accurately and quickly.

  In the eyeglass lens suction jig mounting apparatus according to the embodiment of the present invention, the mark detection means (arithmetic control circuit 170) performs binarization processing of an image acquired by the imaging means (CCD cameras 105 and 106). Then, noise removal processing is performed, and an edge image of the fitting mark or hidden mark on the spectacle lens L is acquired.

  Furthermore, in the spectacle lens suction jig mounting device according to the embodiment of the present invention, the mark detection means (arithmetic control circuit 170) creates an edge from the image acquired by the imaging means (CCD cameras 105 and 106). A hidden mark is detected by applying a template to the edge.

  In the eyeglass lens suction jig mounting apparatus according to the embodiment of the present invention, the position determining means identifies the suction jig mounting position from the position of the acquired hidden mark or fitting mark image. .

  In the eyeglass lens suction jig attachment device according to the embodiment of the present invention, the position determining means identifies the attachment location of the suction jig 121 from the center of gravity of the acquired hidden mark or fitting mark image. ing.

It is a flowchart which shows the whole process of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the process of frame replacement lens holder detection of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a figure which shows the process of frame replacement lens holder detection of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the process of lens presence determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of the process of the lens presence determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the process of lens presence determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of frame replacement lens holder detection of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of the frame replacement lens holder detection of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of the frame replacement lens holder detection of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is the image at the time of the lens detection in mounting | wearing with the frame replacement lens holder of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is the image at the time of the lens detection in mounting | wearing with the frame replacement lens holder of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is the image at the time of the lens detection in mounting | wearing with the frame replacement lens holder of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the procedure of lens discrimination | determination of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the procedure of the bifocal lens discrimination | determination of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of the bifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a figure which shows the gravity center at the time of the bifocal lens determination of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image at the time of the progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an edge detection image at the time of the progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is the image after noise removal at the time of progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image of radius determination at the time of progressive multifocal lens determination of the eyeglass lens suction jig attachment device according to the embodiment of the present invention. It is an image for horizontal line detection at the time of progressive multifocal lens determination of the spectacle lens suction jig mounting device according to the embodiment of the present invention. It is an enlarged image for horizontal line detection at the time of progressive multifocal lens determination of the eyeglass lens suction jig attachment device according to the embodiment of the present invention. It is an image for a hidden mark detection at the time of progressive multifocal lens determination of the eyeglass lens suction jig attachment device according to the embodiment of the present invention. It is an image for a hidden mark detection at the time of progressive multifocal lens determination of the spectacle lens suction jig attachment device according to the embodiment of the present invention. It is an enlarged image for a hidden mark detection at the time of progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an enlarged image for a hidden mark and paint detection at the time of progressive multifocal lens determination of the eyeglass lens suction jig mounting device according to the embodiment of the present invention. It is an enlarged image for the hidden mark at the time of progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention, and a paint gravity center detection. It is a template for a hidden mark detection at the time of progressive multifocal lens determination of the spectacle lens suction jig attachment device according to the embodiment of the present invention. It is an enlarged image for the hidden lens center detection at the time of progressive multifocal lens determination of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the process of the segment detection of the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image explaining the distance function for the segment detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the process of the segment detection of the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the spectacle lens suction jig mounting device according to the embodiment of the present invention. It is an image for the segment detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment detection of the bifocal lens of the adsorption jig attachment device of the spectacle lens concerning an embodiment of the invention. It is a perspective view which shows the state which mounted the bifocal lens in the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the whole process of the segment corner detection of the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image which shows the result of having detected the segment corner with the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a figure which shows the curve fitting of the segment corner detection of the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a flowchart which shows the segment corner detection process of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a figure which shows the curve fitting of the segment corner detection of the bifocal lens of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a graph of the distance function which shows the result of having detected the segment corner with the bifocal lens of the adsorption | suction jig mounting apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image which shows the result of having detected the segment corner with the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for the segment corner detection of the bifocal lens of the adsorption | suction jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is an image for judging the right and left of the progressive multifocal lens of the eyeglass lens suction jig attaching device according to the embodiment of the present invention. It is an image for judgment of the left and right of the bifocal lens of the eyeglass lens suction jig attaching device according to the embodiment of the present invention. It is a block diagram which shows schematic structure of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a perspective view which shows the lens mounting part of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. It is a perspective view which shows the frame replacement lens holder of the adsorption jig attachment apparatus of the spectacle lens which concerns on embodiment of this invention. FIG. 68 is an explanatory diagram illustrating an example of a layout screen of the liquid crystal display of FIG. 67. FIG. 68 is an explanatory diagram showing another example of the layout screen of the liquid crystal display shown in FIG. 67. FIG. 68 is an explanatory diagram illustrating an example of an initial setting screen of the liquid crystal display of FIG. 67. FIG. 68 is an explanatory diagram showing another example of the initial setting screen of the liquid crystal display shown in FIG. 67.

Explanation of symbols

105, 106 ... CCD camera (imaging means)
120 ... Suction jig transport section (part of the mounting means)
121 ... Adsorption jig 140 ... Lens mounting part (mounting table)
141... Opening 150... XY stage (part of mounting means, part of mounting position determining means)
170 ... arithmetic control circuit (part of mark detection means, mounting position determination means, screen display setting means, etc.)
200 ... Lens holder (mounting table)
1503 ... Shadow indicating a hidden mark 1505 ... Shadow indicating an eye point (fitting mark)
L: Eyeglass lenses F1 to F6: Function keys (a part of switching means or screen display setting means)

Claims (6)

A mounting table in which the spectacle lens can be placed in the opening;
Imaging means for taking an image of the spectacle lens placed in the opening;
Position determining means for specifying the attachment position of the suction jig from the imaged image of the spectacle lens;
In a spectacle lens suction jig mounting device comprising a mounting means for placing a suction jig at the mounting position of the spectacle lens,
Said position determining means, mark detecting means for detecting the presence and position of the horizontal reference hidden marks from the opening inside the image captured by the paint eyepoint become hidden marks and the spectacle lens of the spectacle lens in the imaging unit When,
A mounting position determining means for determining a mounting position of the suction jig in the spectacle lens from the detected position;
Switching to switch between a hidden mark, which is an eye point of the spectacle lens, as a reference for the attachment position of the suction jig or a horizontal reference hidden mark by the paint of the spectacle lens as a reference for the attachment position of the suction jig And an eyeglass lens suction jig attaching device. A mounting table that can be placed in the opening of the spectacle lens;
Imaging means for capturing an image of the spectacle lens placed in the opening;
Position determining means for specifying the attachment position of the suction jig from the imaged image of the spectacle lens;
In a spectacle lens suction jig mounting device comprising a mounting means for placing a suction jig at the mounting position of the spectacle lens,
Said position determining means, mark detecting means for detecting the presence and position of the horizontal reference hidden marks from the opening inside the image captured by the paint eyepoint become hidden marks and the spectacle lens of the spectacle lens in the imaging unit When,
A mounting position determining means for determining a mounting position of the suction jig in the spectacle lens from the detected position;
When detected by the mark detection means, the screen display is based on the hidden mark which is the eye point of the spectacle lens, or the screen display is based on the horizontal reference hidden mark by the paint of the spectacle lens. An apparatus for attaching an eyeglass lens suction jig, comprising: a screen display setting means for setting the eyeglass lens. The mark detection unit performs binarization processing and noise removal processing on the image acquired by the imaging unit, and generates an edge image of a hidden mark serving as an eye point on the spectacle lens or a horizontal reference hidden mark by painting on the spectacle lens. The eyeglass lens suction jig mounting device according to claim 1, wherein the eyeglass lens suction jig mounting device is obtained. The mark detection means performs matching between the edge image acquired by the imaging means and a template of a mark prepared in advance to correlate the edge image with the template, and the portion having the highest correlation value is the hidden mark. The apparatus for detecting an eyeglass lens suction jig according to claim 3, wherein: 5. The eyeglass lens suction treatment according to claim 4, wherein the position determining means identifies the attachment position of the suction jig from the position of the acquired hidden mark serving as an eye point or the horizontal reference hidden mark image of the eyeglass lens. Equipment mounting device. The position determining means detects a luminance value of a pixel of the image of the captured spectacle lens, determines that a paint mark is included in the image when the detected luminance value has a pixel of 0, The coordinate average of the portion of the pixel having the luminance value of 0 is obtained as the center of gravity of the paint mark, or when there is a pixel having the detected luminance value of a predetermined value, it is determined that a hidden mark is included in the image. 6. The eyeglass lens suction jig according to claim 5 , wherein a coordinate average of a pixel portion having a predetermined luminance value is obtained as a center of gravity of a hidden mark, and an attachment position of the suction jig is specified from the center of gravity. Mounting device.

Images