METHOD FOR OPTICALLY INSPECTING AN OPHTHALMIC LENS TO DETERMINE AN ORIENTATION STATE AND AN INVERSION STATE OF AN OPHTHALMIC LENS

20230304890 · 2023-09-28

    Inventors

    Cpc classification

    International classification

    Abstract

    A method for optically inspecting an ophthalmic lens in an automated lens production process to determine an orientation state and an inversion state of the ophthalmic lens. The method comprises the steps of acquiring an image of the ophthalmic lens, determining from the acquired image an orientation value of the lens, comparing the determined orientation value with an orientation threshold value, and determining that the lens is oriented right side up when the determined orientation value is higher than the orientation threshold value, or determining that the lens is oriented upside down when the determined orientation value is lower than the orientation threshold value, or vice versa.

    Claims

    1. Method (100) for optically inspecting an ophthalmic lens (2) in an automated lens production process to determine an orientation state and an inversion state of the ophthalmic lens (2), the method comprising the steps of acquiring (101) an image of the ophthalmic lens (2), determining (102) from the acquired image an orientation value of the ophthalmic lens (2), comparing (103) the determined orientation value with an orientation threshold value, and determining that the ophthalmic lens is oriented right side up (104b) when the determined orientation value is higher than the orientation threshold value, or determining that the ophthalmic lens is oriented upside down (104a) when the determined orientation value is lower than the orientation threshold value, or vice versa, determining from the acquired image an inversion value of the ophthalmic lens (2), comparing the determined inversion value (105a, 105b) of the optically inspected ophthalmic lens (2) with an inversion threshold value and determining that the lens is non-inverted (108a, 108b) when the determined inversion value is higher than the inversion threshold value, or determining that the ophthalmic lens is inverted (107a, 107b) when the determined inversion value is lower than the inversion threshold value, or vice versa, wherein at least one of the orientation threshold value and the inversion threshold value is determined based on a distribution of orientation values and a distribution of inversion values of previous optically inspected ophthalmic lenses, respectively, characterized in that the at least one of the orientation threshold value and the inversion threshold value is dynamically adapted during the automated lens production process.

    2. Method according to claim 1, wherein both the orientation threshold value and the inversion threshold value are determined based on the distribution of orientation values and the distribution of inversion values of previous optically inspected ophthalmic lenses (2), respectively, and wherein both the inversion threshold value and the orientation threshold value are dynamically adapted during the automated lens production process.

    3. Method according to claim 1, wherein determining the orientation state of the ophthalmic lens (2) is performed prior to determining the inversion state of the ophthalmic lens (2).

    4. Method according to claim 3, wherein the ophthalmic lens (2) is an ophthalmic lens (2) containing coloring pigments, wherein determining the orientation value is performed by determining the degree of defocus of a central portion (4, 5) of the image of the ophthalmic lens, and wherein for the ophthalmic lens (2) determined to be oriented right side up the inversion value is then determined by calculating a deformation image noise level (105b) related to a deformation state of the coloring pigments of a section (61) of the image of the ophthalmic lens, whereas for the ophthalmic lens determined to be oriented upside down the inversion value is then determined by determining (105a) a width of a dark edge zone in the image of the ophthalmic lens.

    5. Method according to claim 4, wherein determining the degree of defocus of the central portion (4, 5) of the image of the ophthalmic lens containing the coloring pigments comprises calculating a defocus image noise level related to an amount of blur of the central portion of the image.

    6. Method according to claim 2, wherein the dynamically adapted orientation threshold value and the dynamically adapted inversion threshold value are determined during the automated production process based on a distribution of orientation values and a distribution of inversion values, respectively, of a predetermined number of the most recent previous optically inspected ophthalmic lenses.

    7. Method according to claim 6, wherein the predetermined number of the most recent previous optically inspected ophthalmic lenses is in a range of 100 to 10000.

    8. Method according to claim 7, wherein the predetermined number of the most recent previous optically inspected ophthalmic lenses in a range of 1000 to 3000.

    9. Method according to claim 4, wherein the distribution of orientation values of the most recent previous optically inspected ophthalmic lenses comprises a lower range (75) of orientation values of the most recent previous optically inspected ophthalmic lenses oriented upside down, and a higher range (76) of orientation values of the most recent previous optically inspected ophthalmic lenses oriented right side up, wherein the dynamically adapted orientation threshold value is determined to be an arithmetic mean value (72) of a highest value (73) of the lower range of orientation values and a lowest value (74) of the higher range of orientation values, wherein for the ophthalmic lenses oriented upside down, the distribution of inversion values comprises a distribution of values of the width of the dark edge zone in the images of the ophthalmic lenses comprising a lower range (95) of values of the width of the dark edge zone in the images of the inverted ophthalmic lenses, and a higher range (96) of values of the width of the dark edge zone in the images of the non-inverted ophthalmic lenses, wherein the dynamically adapted inversion threshold value is determined to be an arithmetic mean value (92) of a highest value (93) of the lower range of values and a lowest value (94) of the higher range of values of the width of the dark edge zone, whereas for the ophthalmic lenses oriented right side up, the distribution of inversion values comprises a distribution of deformation image noise level values comprising a lower range (85) of deformation image noise level values of the inverted ophthalmic lenses, and a higher range (86) of deformation image noise level values of the non-inverted ophthalmic lenses, and wherein the dynamically adapted inversion threshold value is determined to be an arithmetic mean value (82) of a highest value (83) of the lower range of deformation image noise level values and a lowest value (84) of the higher range of deformation image noise level values.

    10. An ophthalmic lens inspection system (1) configured to perform the method according to claim 1, the system comprising: an imaging unit (10) for acquiring an image of the ophthalmic lens (2), a processor (17) configured to perform the steps of: determining the orientation value from the image of the ophthalmic lens (2) acquired by the imaging unit, comparing the determined orientation value with the orientation threshold value, determining that the ophthalmic lens is oriented right side up, or determining that the ophthalmic lens is oriented upside down, determining the inversion value from the image of the ophthalmic lens (2) acquired by the imaging unit, comparing the determined inversion value with the inversion threshold value, determining that the ophthalmic lens (2) is non-inverted, or determining that the ophthalmic lens is inverted, and determining the at least one of the orientation threshold value and the inversion threshold value based on the distribution of orientation values and the distribution of inversion values of previous optically inspected ophthalmic lenses, respectively, wherein the processor (17) is further configured to dynamically adapt the at least one of the inversion threshold value and the orientation threshold value during the automated lens production process.

    11. The ophthalmic lens inspection system (1) according to claim 10, wherein the imaging unit comprises a camera (14) for acquiring the image of the ophthalmic lens (2).

    Description

    BRIEF DESCRIPTION OF THE DRAWINGS

    [0065] Further advantageous aspects of the invention become apparent from the following description of embodiments of the invention with the aid of the (schematic) drawings, in which:

    [0066] FIG. 1 shows components of an embodiment of the ophthalmic lens inspection system according to the invention;

    [0067] FIG. 2 shows a non-inverted soft contact lens oriented right side up and arranged on a lens holder;

    [0068] FIG. 3 shows an inverted soft contact lens oriented right side up and arranged on the lens holder;

    [0069] FIG. 4 shows a non-inverted soft contact lens oriented upside down and arranged on the lens holder;

    [0070] FIG. 5 shows an inverted soft contact lens oriented upside down and arranged on the lens holder;

    [0071] FIG. 6 shows a focused image of a central portion of a non-inverted soft contact lens oriented right side up;

    [0072] FIG. 7 shows a defocused image of a central portion of a non-inverted soft contact lens oriented upside down;

    [0073] FIG. 8 shows a focused image of a section of an inverted soft contact lens oriented right side up;

    [0074] FIG. 9 shows images of four soft contact lenses having different orientation and inversion states;

    [0075] FIG. 10 shows a flowchart of an embodiment of the method according to the invention;

    [0076] FIG. 11 shows a histogram of a distribution of orientation values of the most recent ones of previous optically inspected soft contact lenses;

    [0077] FIG. 12 shows a histogram of a distribution of deformation image noise values of the most recent ones of previous optically inspected soft contact lenses oriented right side up; and

    [0078] FIG. 13 shows a histogram of a distribution of values of the width of a dark edge zone in the images of the most recent ones of previous optically inspected soft contact lenses oriented upside down.

    [0079] FIG. 1 shows schematically an embodiment of components of an ophthalmic lens inspection system 1 according to the invention, in particular an imaging unit 10. In FIG. 1, a container 11 accommodating a soft contact lens 2 to be optically inspected is arranged in the imaging unit 10. The container 11 is completely filled with a liquid such as water or saline, in which the soft contact lens 2 is immersed. The container 11 comprises a container bottom 12 having a concave upper container bottom surface 13 for the soft contact lens 2 to rest on. The container bottom 12 is made of a transparent material such as quartz glass, and the concave upper container bottom surface 13 as well as a lower container bottom surface 18 of the container bottom 12 are polished to optical quality. The lower container bottom surface 18 is convexly shaped such that the container bottom 12 has light focusing properties. The imaging unit 1 comprises a light source 19 and a camera 14. Light is irradiated from the light source 19 through the container bottom 12 and the soft contact lens 2 resting on the concave upper container bottom surface 13 allowing the camera 14 to acquire an image of the soft contact lens 2. The camera 14 comprises an objective 16 for focusing the light transmitted through the container bottom 12 and the soft contact lens 2 onto a light sensor 15 such as a CCD-sensor or a CMOS-sensor. The objective 16 may comprise a telecentric objective. The imaging unit may comprise further optical elements such as focusing elements and mirrors which are not shown in the drawing.

    [0080] The image of the soft contact lens 2 is then further processed by a processor 17 of the ophthalmic lens inspection system. The processor 17 is configured to perform the necessary computational steps to determine an orientation state of the soft contact lens 2 from the acquired image, and to determine an inversion state of the soft contact lens 2 either from the acquired image, as is described in more detail further below.

    [0081] FIG. 2 shows the soft contact lens 2 arranged on a lens holder 3. The lens holder 3 may be formed by the container bottom 12 shown in FIG. 1, however, due to the shape of the lens holder 3 shown in FIG. 2 deviating from the entirely concave shape of the container bottom 12 shown in FIG. 1 it has been assigned a different reference sign in FIG. 2. The lens holder 3 has a concave surface 30 for the soft contact lens 2 to rest on for the optical inspection. For the avoidance of doubt, this optical inspection of the soft contact lens 2 is performed for the purpose of determining the orientation of the soft contact lens 2 and the inversion state of the soft contact lens, and has nothing to do with the inspection of the soft contact lens 2 for cosmetic defects such as scratches, inclusions, bubbles, tears or edge defects which is performed only once the soft contact lens 2 is oriented right side up and is non-inverted. The soft contact lens 2 comprises a convexly shaped front surface 24 and a concavely shaped rear surface 23 which is in contact with the eye of the wearer when the soft contact lens 2 is worn. The soft contact lens 2 shown in FIG. 2 is oriented right side up, such that the convexly shaped front surface 24 of the soft contact lens 2 faces the concave surface 30 of the lens holder 3. In this case a central portion 20 of the soft contact lens 2 is within a depth of field 26 of the camera 14. Only portions of the soft contact lens which are within the depth of field 26 appear focused in an image of the contact lens 2. The depth of field 26 may extend horizontally as shown in FIG. 2, but it may in an alternative embodiment also extend as a curved shape following the shape of the concave surface 30 of the lens holder 3. The soft contact lens 2 shown in FIG. 2 is non-inverted.

    [0082] FIG. 3 shows the soft contact lens 2 arranged on the lens holder 3, however, with the soft contact lens 2 being inverted (i.e. in a different inversion state than in FIG. 2). The soft contact lens 2 is again oriented right side up, such that a convex surface of the inverted soft contact lens 2 faces the concave surface 30 of the lens holder 3. However, due to the soft contact lens 2 being inverted the convex surface of the soft contact lens 2 facing the concave surface 30 of the lens holder 3 is the rear surface 23 of the soft contact lens 2 which is normally concavely shaped. The central portion 20 of the soft contact lens 2 in this inversion state and orientation state (i.e. oriented right side up and inverted) is again within the depth of field 26 of the camera 14.

    [0083] FIG. 4 shows the soft contact lens 2 arranged on the lens holder 3, however with the soft contact lens 2 being oriented upside down. The soft contact lens 2 shown in FIG. 3 is non-inverted. In this orientation and inversion state, the rear surface 23 of the soft contact lens 2 is concavely shaped and the front surface 24 is convexly shaped, as desired. However, due to the contact lens 2 being oriented upside down the concavely shaped rear surface 23 faces the concave surface 30 of the lens holder 3. In this orientation state and inversion state, the central portion 20 of the soft contact lens 2 is outside the depth of field 26 of the camera 14.

    [0084] FIG. 5 shows the soft contact lens 2 arranged on the lens holder 3, however with the soft contact lens 2 being oriented upside down. The soft contact lens shown in FIG. 4 is inverted. In this state, the front surface 24—which in this state is concavely shaped-faces the concave surface 30 of the lens holder 3. Also, in this state of the soft contact lens 2, the central portion of the soft contact lens 2 is outside the depth of field 26 of the camera 14.

    [0085] FIG. 6 shows a focused central portion 4 of an image of the non-inverted soft contact lens 2 oriented right side up (state of the soft contact lens 2 shown in FIG. 2). The focused central portion 4 of the image contains a plurality of sharply imaged coloring pigments 40 (only three of them being labelled with reference sign 40 in FIG. 6). The coloring pigments are contained in a structural pattern in the soft contact lens 2 such as described in US 2019/0072784 or WO 2015/036432. As the central portion 20 of the soft contact lens 2 is arranged within the depth of field 26 of the camera 14, features of the central portion 20 of the contact lens 2 are in focus in the image. Thus, the imaged coloring pigments 40 appear unblurred (i.e. with a low degree of defocus).

    [0086] FIG. 7 shows a defocused central portion 5 of an image of the non-inverted soft contact lens 2 oriented upside down (state of the soft contact lens 2 shown in FIG. 4). The defocused central portion 5 contains a plurality of blurredly imaged coloring pigments 50 (again only three of them being labelled with reference sign 50 in FIG. 7), whereas the coloring pigments contained in the soft contact lens 2 are the same as the ones visible as sharply imaged coloring pigments in the focused central portion 4 of the image of the soft contact lens 2 oriented right side up (see FIG. 6). However, in contrast to FIG. 6, the central portion 20 of the soft contact lens 2 is arranged outside the depth of field 26 of the camera 14 (see FIG. 4). Thus, features of the central portion 20 of the soft contact lens 2, such as the blurredly imaged coloring pigments 50 are not in focus in the image. Thus, the imaged coloring pigments 50 appear blurred (i.e. with a high degree of defocus) in the defocused central portion 5 of the image.

    [0087] As can be seen from FIG. 6 and FIG. 7, the degree of defocus of the central portion 4, 5 of the image may well be used to determine the orientation state of the soft contact lens 2 (i.e. right side up or upside down).

    [0088] FIG. 10 shows a flowchart illustrating an embodiment of the method 100 according to the invention to determine both the orientation state and the inversion state of a soft contact lens 2. In step 101, an image of the soft contact lens 2 is acquired. As already discussed above in connection with FIG. 6-FIG. 8, the image of the soft contact lens 2 may be a telecentric bright field image acquired by the camera 14. In step 102, a defocus image noise level is then determined from a central portion of the image as described in US 2019/0072784, this defocus image noise level representing the orientation value. As can be seen from FIG. 6 and FIG. 7, the grey scale values of the focused central portion 4 of the image (see FIG. 6) have a stronger variation compared to the grey scale values of the defocused central portion 5 of the image (see FIG. 7). The defocus image noise level can therefore be determined as the standard deviation of the grey scale values of the respective central portion 4, 5 of the image of the soft contact lens 2. Filters may be applied to the images for the determination of the standard deviation of the grey scale values. Such filters may comprise a Wiener filter or a Fourier transform, as described in US 2019/0072784. Moreover, an edge detection filter may be applied prior to the calculation of the said standard deviation. The so obtained value for the standard deviation is then used as the orientation value and is compared with a dynamically adapted orientation threshold value in step 103. The soft contact lens 2 is determined to be oriented right side up when the orientation value is higher than the dynamically adapted orientation threshold value, and is determined to be oriented upside down when the orientation value is lower than the dynamically adapted orientation threshold value. Thus, the soft contact lens 2 is determined as either being oriented upside down, as illustrated by step 104a, or it is determined as being oriented right side up, as illustrated by step 104b. Steps 102, 103, 104a, 104b of the method are performed by the processor 17 (FIG. 1).

    [0089] The manner how the inversion state of the soft contact lens 2 is determined depends on the orientation state of the contact lens that is already determined as described above (lens oriented right side up or upside down).

    [0090] For the soft contact lens 2 determined to be oriented right side up (step 104b), the inversion state can be determined with the aid of the coloring pigments contained in the soft contact lens 2 which are contained as imaged coloring pigments in the image of the soft contact lens 2. FIG. 8 shows a focused central portion 6 of an image of an inverted soft contact lens 2 oriented right side up. Due to the soft contact lens 2 being oriented right side up, the central portion 20 of the soft contact lens 2 is within the depth of field 26 of the camera 14. However, in an alternative embodiment the portion of the soft contact lens 2 that is in focus may be a portion other than the central portion 20, as long as the said portion is within the depth of field 26 of the camera 14. As described in WO 2015/036432, the coloring pigments contained in the inverted soft contact lens 2 are deformed to line structures. The central portion 6 of the image comprises a plurality of imaged deformed coloring pigments 60. Thus, the inversion state of the soft contact lens 2 can be determined with the aid of the imaged deformed coloring pigments 60 contained in the image. When the coloring pigments are getting deformed, the variation of grey scale values of the imaged pigments 60 contained in the focused central portion 6 of the image is reduced (due to the pigments being stretched). A deformation image noise level can then be determined in a manner similar to that described for the non-inverted soft contact lens 2, this deformation image noise level forming the inversion value representing the inversion state of the soft contact lens 2. That is, determining the deformation image noise level in step 105b (FIG. 10) is performed by determining a standard deviation of the grey scale values of a section 61 of the image (FIG. 8). This deformation image noise level is compared with a dynamically adapted inversion threshold value, as illustrated by step 106b (FIG. 10). In case the deformation image noise level is higher than the dynamically adapted inversion threshold value the soft contact lens 2 is determined to be non-inverted, as illustrated by step 108b, and in case the deformation image noise level is lower than the inversion threshold value the contact lens 2 is determined to be inverted, as illustrated by step 107b. In alternative embodiment, the presence of the imaged deformed coloring pigments 60 can be determined by detecting and counting the imaged deformed coloring pigments 60 in section 61 as described in prior art WO 2015/036432.

    [0091] For the soft contact lens 2 determined to be oriented upside down (step 104a), the inversion state of the soft contact lens 2 is determined from the width 250 of a dark edge zone 251 in the image of the soft contact lens 2. FIG. 9 schematically shows images of four soft contact lens 2 in four different orientation and inversion states: [0092] the soft contact lens 2 is oriented upside down and is non-inverted (upper left image) [0093] the soft contact lens 2 is oriented right side up and is non-inverted (upper right image) [0094] the soft contact lens 2 is oriented upside down and is inverted (lower left image) [0095] the soft contact lens 2 is oriented right side up and is inverted (lower right image).

    [0096] As can be seen, when the soft contact lens 2 is in a non-inverted state (upper right and upper left images) the width 251 of a dark edge zone 250 extending radially inwardly from the outermost boundary or edge 25 in the image of the soft contact lens 2 is substantially larger than the width 251 of the dark edge zone 250 in the image of the soft contact lens 2 when the lens is in an inverted state (lower left and lower right images). Accordingly, the width 251 of the dark edge zone 250 can be used to determine the inversion state of the soft contact lens, i.e. to determine whether or not the soft contact lens 2 is inverted. The determined width 251 of the dark edge zone 250 is then compared with a dynamically adapted inversion threshold value, as is illustrated by step 105a (FIG. 10), this width 251 of the dark edge zone 250 forming the inversion value representing the inversion state of the soft contact lens. This width 251 of the dark edge zone 250 is then comparted with a dynamically adapted inversion threshold value, as illustrated by step 106a. In case the determined width 251 is larger than the dynamically adapted threshold value, the soft contact lens 2 is determined to be non-inverted, as illustrated by step 108a, and in case the width 251 is lower than the dynamically adapted inversion threshold value, the soft contact lens 2 is determined to be inverted, as illustrated by step 107a. Steps 105a, 105b, 106a, 106b, 107a, 107b, as well as steps 108a and 108b of the method are again performed by the processor 17 (FIG. 1).

    [0097] The dynamically adapted orientation threshold value may be determined from a distribution of orientation values of the most recent optically inspected soft contact lenses 2, in particular of the values of the defocus image noise level of the said lenses. The orientation values of a number in the range of 100 to 10000, in particular in the range of 1000 to 3000, of the most recent ones of the previous optically inspected contact lenses may be considered when determining the dynamically adapted orientation threshold value (an update of the histograms may be performed periodically, e.g. after every ten or every hundred soft contact lenses 2 have been inspected). A histogram 7 of such a distribution of orientation values of the most recent ones of the previous optically inspected soft contact lenses 2 is shown in FIG. 11. The horizontal axis 71 (abscissa) is the axis of the orientation values, with the defocus image noise levels increasing from left to right. The vertical axis 70 (ordinate) is the axis of the frequency of the values having the respective defocus image noise level. The orientation values accumulate in a lower range 75 of orientation values and a higher range 76 of orientation values. The orientation threshold value 72 is determined as an arithmetic mean of the highest orientation value 73 of the lower range 75 of orientation values and the lowest orientation value 74 of the higher range 76 of orientation values.

    [0098] To determine the dynamically adapted inversion threshold value for the soft contact lenses 2 oriented right side up, a distribution of the values of the deformation image noise level of the most recent ones of the previous optically inspected soft contact lenses 2 oriented right side up is considered. These previous optically inspected soft contact lenses 2 oriented right side up are highlighted by the rectangle XII in FIG. 11. A histogram 8 of the said distribution is shown in FIG. 12. The distribution of the deformation image noise level values again comprises a lower range 85 of deformation image noise level values and a higher range 86 of deformation image noise level values. The inversion threshold value for the soft contact lenses 2 oriented right side up is then determined as an arithmetic mean value of the highest deformation image noise value 83 of the lower range 85 of deformation image noise level values and the lowest deformation image noise value 84 of the higher range 86 of deformation image noise level values.

    [0099] To determine the dynamically adapted inversion threshold value for the soft contact lenses oriented upside down, a distribution of values of the edge thickness of the most recent ones of the previous optically inspected soft contact lenses 2 oriented upside down is considered. These previous optically inspected soft contact lenses 2 oriented upside down are highlighted by the rectangle XIII in FIG. 11. A histogram 9 of the said distribution is shown in FIG. 13. The distribution of edge thickness values comprises a lower range 95 of edge thickness values and a higher range 96 of edge thickness values. The inversion threshold value for the soft contact lenses 2 oriented upside down is then determined as an arithmetic mean value of the highest edge thickness value 93 of the lower range 95 of edge thickness values and the lowest edge thickness value 94 of the higher range 96 of edge thickness values.

    [0100] It is evident, that the afore-described orientation threshold value and inversion threshold value may be dynamically adapted during the automated lens production process from the data of the most recent ones of the previous optically inspected soft contact lenses 2. Common production process variations such as changes in temperature, raw material batches or lens design, that may lead to changes of the orientation values and inversion values determined from the optical inspection of soft contact lenses 2 having the same orientation state and inversion state, can be accounted for through the dynamic adaptation of the orientation threshold value and the inversion threshold value. Thus, the determination of the orientation state of the soft contact lenses 2 (right side up or upside down) as well as the determination of the inversion state of the soft contact lenses 2 (inverted or non-inverted) can be made more reliably during the running production even when common production variations occur, thus increasing the production yield.

    [0101] While embodiments of the invention have been described with the aid of the drawings, the invention is not limited to these embodiments, but rather various changes and modifications are possible without departing from the teaching underlying the invention. Therefore, the scope of protection is not intended to be limited to the embodiments described, but rather is defined by the appended claims.