JP2014228809A - Compound-eye optical system, image-capturing device, and method for manufacturing compound-eye optical system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a compound-eye optical system capable of reducing crosstalk or ghost due to light propagating inside a lens array, and a manufacturing method therefor.SOLUTION: In a first lens array 10 constituting a compound-eye optical system 100, since first patterns PT1 on an optical surface 11a side of an image-side principal surface face substantially the same direction, light refracted by a specific lens element to reach an adjacent lens element is repeatedly refracted by the first patterns PT1 directed in the same direction. Due to such repeated refraction by the first patterns PT1, stray light which is the cause of crosstalk can be reduced, and the performance of the compound-eye optical system 100 can be improved.

Description

  The present invention relates to a compound eye optical system, an imaging device incorporating the compound eye optical system, and a method for manufacturing the compound eye optical system.

In recent years, there is a great demand for thinning the imaging optical system. To cope with this, it has been attempted to shorten the overall length by optical design and improve manufacturing accuracy to cope with the accompanying increase in error sensitivity. It has become difficult to meet the further demands of the conventional method of obtaining.
Therefore, the imaging device is divided into a plurality of regions, an optical system is arranged in each region, and a final image output is performed by processing the obtained image, and a compound eye optical system used therefor The so-called optical system has attracted attention from the viewpoint of meeting the demand for thinning.

  Various compound-eye optical systems have been proposed to date, but a thorough examination of techniques for achieving high image quality and ultra-thinness using optical resin materials such as thermoplastic resins that are lightweight and excellent in cost reduction There are no examples.

  The compound eye optical system is premised on a plurality of independent single eyes, and it is preferable to provide an interval between the single eyes in order to avoid interference of a plurality of images obtained by each single eye. In particular, when the compound eye optical system is made of resin, it is preferable to provide a flange portion outside the effective diameter of each lens portion from the viewpoint of improving molding conditions for obtaining a lens array by injection molding. However, according to the study by the inventors of the present application, it has been found that light propagates through the flange portion and crosstalk occurs, which may be observed as a ghost. That is, there is a possibility that a ghost is generated when light incident from a certain eye propagates to an adjacent eye via a flange portion.

Patent Documents 1 and 2 describe a compound eye imaging device using a resin lens array having a flange portion.
Further, in the prior art column of Patent Document 3, a plastic optical element in which lens elements are arranged in a grid shape without a gap is disclosed as a lens array. This optical element is a molded product obtained by side gate type injection molding.
In addition, although it does not relate to a compound eye optical system, it relates to a single lens system, but in Patent Document 4, by changing the orientation of a large number of linear fine grooves formed on the resin lens surface by 90 degrees on the front and back, It is described that the influence of wavefront aberration is reduced.

However, Patent Documents 1 and 2 describe a structure in which a light shielding wall is disposed between the image sensor and the lens array to prevent crosstalk, and a diaphragm is disposed on the object side of the lens array. Crosstalk due to light propagating in the array is not considered.
In the prior art column of Patent Document 3, there is an example of a plastic lens array of side gates, but this is for an image display device, and ghost is not mentioned.
Since Patent Document 4 relates to the reduction of wavefront aberration of a light beam transmitted through the entrance surface and the exit surface of a single lens, it does not describe reduction of ghosts between a plurality of adjacent lenses.

JP 2007-94013 A JP 2011-147079 A JP-A-8-248207 JP 2006-91328 A

The present invention has been made in view of the above-described background art, and an object thereof is to provide a compound eye optical system capable of reducing crosstalk or ghost caused by light propagating in an array and a method for manufacturing the same.
Another object of the present invention is to provide an imaging apparatus incorporating the compound eye optical system.

  In order to solve the above-described problems, the compound eye optical system according to the present invention is an integral body made of an optical material including a plurality of lens elements arranged two-dimensionally and a flange portion extending between the plurality of lens elements. A lens array is provided, and on the image-side main surface of the two main surfaces of the lens array, a first pattern composed of periodic irregularities extending linearly is formed on a plurality of first optical surfaces of a plurality of lens elements. The first patterns of the plurality of first optical surfaces are directed in substantially the same direction.

In the above compound-eye optical system, the first pattern faces substantially the same direction on the image side main surface, so that it enters the lens element from the object side, is diffracted and reflected by the image side optical surface, propagates through the flange portion, and is adjacent to the lens element. The stray light that causes crosstalk can be reduced by the combined diffraction of the light reaching the image side optical surface of the lens element by the first pattern of the image side optical surfaces of two adjacent lenses. Here, the periodic pattern is formed by transferring a tool mark or the like formed on the transfer surface of the mold.
The tool mark is a non-flat trace that occurs accompanying the formation of the transfer surface by the processing tool. The tool mark is usually a periodic pattern, and it is possible to erase the tool mark by polishing the optical transfer surface of the mold, but there is a possibility that the surface accuracy may be lowered due to shape variation in the erasing process. For this reason, in applications where surface accuracy is required, it is necessary to allow the tool marks to remain to some extent. Even when the tool mark is left in this way, the direction of the first pattern in which the shape of the tool mark is reflected is matched between two lenses adjacent to each other on the image side main surface. Further, the light that is totally reflected by the flange portion and reaches the optical surface on the image side of the adjacent lens element is repeatedly diffracted by the first pattern. As a result, unintentional rays incident on a specific lens element can be prevented from entering the optical path of the adjacent lens element with a relatively large intensity due to diffraction, and the phenomenon such as image detection is deteriorated due to the crosstalk of the light. Can be suppressed.

  In a specific aspect or embodiment of the present invention, in the above compound eye optical system, the lens array is molded by injecting a thermoplastic resin into a molding die, and a resin injection portion is provided on the side of the outer edge connecting the two main surfaces. It has been. The lens array can be molded with a thermoplastic resin at a relatively low cost. In particular, when a thermoplastic resin is injected into the mold space via a gate corresponding to the resin injection portion, the resin can be easily supplied to a portion corresponding to a thin portion of the lens array, and the resulting lens array is relatively high. It can be accuracy. When the resin injection portion is formed on the side portion, the contraction of the molded product is not axisymmetric about the central axis, so it is desirable to make the optical surface or the like a free-form surface in consideration of correction. The transfer mold corresponding to such a free-form surface can easily and highly accurately process the machining tool by moving the machining tool linearly. As a result, the tool mark and the pattern on the optical surface reflecting the tool mark are also periodic. Becomes a straight line.

In still another aspect of the present invention, when the uneven pitch of the first pattern is TP (μm) and the use center wavelength of the plurality of lens elements is λ (μm), the following relationship 1.5 <TP /Λ<8.0
Meet. When the value TP / λ is smaller than 8.0, it is possible to prevent the pitch from becoming rough and the surface accuracy from being lowered. On the other hand, when the value TP / λ is larger than 1.5, it is possible to prevent the pitch from becoming fine and diffracting from occurring, or the problem that specific very strong diffraction is generated can be prevented.

In still another aspect of the present invention, when the height of the unevenness of the first pattern is H (μm) and the use center wavelength of the plurality of lens elements is λ (μm), the following relationship 0.5 <100 × H / λ <7.0
Meet. By setting the height of the periodic pattern so that the value 100 × H / λ is smaller than 7.0, the diffraction efficiency is prevented from becoming too high, and the crosstalk is prevented from increasing beyond the amount that can be reduced. it can. By making the periodic pattern height such that the value 100 × H / λ is larger than 0.5, the diffraction efficiency is prevented from becoming too low, and the crosstalk reduction effect due to the complex diffraction of the periodic pattern Can be secured.

  In still another aspect of the invention, the plurality of lens elements are arranged in a matrix along four orthogonal reference directions.

  In yet another aspect of the present invention, on the object-side main surface of the lens array, a plurality of second optical surfaces of the plurality of lens elements are formed with a second pattern composed of periodic irregularities extending linearly, The second patterns of the plurality of second optical surfaces face substantially the same direction as the first pattern. In this case, since the first pattern on the image side and the second pattern on the object side face in substantially the same direction, the stray light that causes crosstalk is reduced by the combined diffraction of the first pattern and the second pattern. And the performance of the compound eye optical system can be improved.

  In still another aspect of the present invention, a plurality of lens arrays including the lens array described above are stacked. In this case, imaging performance can be improved by combining lens elements of different lens arrays.

  In yet another aspect of the present invention, the plurality of lens arrays includes a first lens array in which the lens element of the lens array described above is a first lens element and the flange portion is a first flange portion. A second lens array that is a single piece including a plurality of second lens elements arranged in a shape and a second flange portion extending between the plurality of second lens elements. A third pattern, which is a periodic unevenness extending linearly, is formed on a plurality of third optical surfaces corresponding to the plurality of second lens elements on at least a main surface on the image side of the main surfaces, and the plurality of third opticals The third pattern of the surfaces faces substantially the same direction.

  In still another aspect of the present invention, the fourth surface which is a periodic unevenness extending linearly to a plurality of fourth optical surfaces corresponding to a plurality of second lens elements on the object-side main surface of the second lens array. A pattern is formed, and the fourth patterns of the plurality of fourth optical surfaces face substantially the same direction as the third pattern.

  In still another aspect of the present invention, the first and second patterns of the first lens array and the third and fourth patterns of the second lens array are oriented in substantially the same direction.

  In still another aspect of the present invention, the first and second patterns of the first lens array and the third and fourth patterns of the second lens array face different directions.

  In yet another aspect of the present invention, a fifth pattern made of periodic irregularities extending linearly on the surface of the flange portion is formed on at least one of the two main surfaces of the lens array that need not be laminated. In this case, since a periodic pattern is formed in addition to the optical surface, an effect of further reducing the intensity of light propagating in the lens array can be expected.

  In still another aspect of the present invention, the first lens array includes periodic irregularities extending linearly on the surface of the flange portion in at least one of the two main surfaces of the first lens array and the two main surfaces of the second lens array. Five patterns are formed.

  In order to solve the above problems, an imaging apparatus according to the present invention includes the compound eye optical system described above and a sensor array having a plurality of sensor elements provided corresponding to the plurality of lens elements.

  Since the imaging apparatus incorporates a compound eye optical system having the characteristics already described, crosstalk can be reduced, and performance can be improved when detecting images in the field division method or super-resolution method. Here, the field division method refers to a method of obtaining one image by joining images of different fields of view formed by individual lenses and connecting the images of the fields of view by image processing. The super-resolution method refers to a method of obtaining one high-resolution image by image processing from images of the same field of view formed by individual lenses.

  In order to solve the above problems, a method for manufacturing a compound eye optical system according to the present invention includes a molding space having a molding surface corresponding to a plurality of lens elements and a flange portion extending between the plurality of lens elements. A first pattern consisting of periodic irregularities extending in a straight line, each of which is included in a plurality of lens elements and faces substantially the same direction, in the step of filling with a lens material and the main surface on the image side of the two main surfaces Obtaining a lens array having a plurality of first optical surfaces.

  In the compound eye optical system obtained by the above manufacturing method, since the first pattern faces substantially the same direction, stray light that causes crosstalk can be reduced by being repeatedly diffracted by the first pattern.

  In a specific aspect of the present invention, the molding surface has tool marks made of periodic irregularities extending linearly at portions corresponding to the plurality of first optical surfaces, and the shape of the tool mark is transferred to the lens material. After that, by removing the molded product from the mold, a lens array including a plurality of integrally formed lens elements and a flange portion is obtained. In this case, the tool mark formed on the molding surface of the mold is transferred to the optical surface of the lens array, which is a molded product, so that a periodic pattern is formed. Simplified.

  In another aspect of the present invention, the tool mark is formed on the molding surface of the mold by raster processing. In this case, the optical surface of the lens array can be processed as a free curved surface.

  In still another aspect of the present invention, the method further includes a step of machining a molding surface of the mold, and the machining step includes a machining data correction step of correcting a non-axisymmetric error by feedback. In this case, it is possible to correct the shape in consideration of shrinkage of the molded product.

(A) And (B) is the top view and side sectional drawing explaining the compound-eye optical system etc. of 1st Embodiment. (A) is a perspective view of the compound eye optical system shown to FIG. 1 (A) etc., (B) is a partially broken perspective view of the compound eye optical system shown to (A). It is sectional drawing which illustrates notionally generation | occurrence | production of the crosstalk by diffraction. It is a figure explaining generation | occurrence | production of the diffracted light by a periodic pattern. (A) is a figure explaining the periodic pattern formed in the optical surface of the lens array of embodiment, (B) demonstrates the periodic pattern formed in the optical surface of the lens array of a reference example. FIG. (A) is a figure explaining notionally the reduction of crosstalk light by diffraction in the compound eye optical system of an embodiment, and (B) conceptually shows crosstalk light by diffraction in the compound eye optical system of a reference example. It is a figure explaining. (A) is sectional drawing explaining the modification of a compound eye optical system, (B) is a conceptual diagram. (A) is sectional drawing explaining another modification of a compound eye optical system, (B) is a conceptual diagram. FIGS. 3A and 3B are a partially enlarged sectional view and a sectional conceptual view for explaining a molding die for a lens array shown in FIG. It is a figure explaining manufacturing methods, such as a compound eye optical system. It is a figure explaining the processing apparatus of the metal mold | die for shaping | molding. (A)-(D) is a figure explaining manufacture of a lens array etc. in steps. (A) is a figure which shows an example of the molded article by injection molding, (B) is a figure which shows another example of the molded article by injection molding. (A) And (B) is a figure explaining the lens array used for the compound-eye optical system of 2nd Embodiment.

[First Embodiment]
As shown in FIGS. 1 (A), 1 (B), 2 (A), and 2 (B), the compound-eye optical system 100 of this embodiment includes a plurality of lens arrays 10 and 20 and a plurality of spacers 30 and 40. It is the laminated body which piled up alternately. In the compound eye optical system 100, the first and second lens arrays 10 and 20 are flat members extending in parallel with the XY plane, and the first and second spacers 30 and 40 are the first and second lens arrays. 10 and 20, and these members 10, 20, 30, and 40 are stacked in the Z-axis direction perpendicular to the XY plane.

  In the compound-eye optical system 100, the first lens array 10 on the object side is an integral body made of a thermoplastic resin, which is an optical material, and has a square or rectangular outline in plan view. The first lens array 10 includes a plurality of lens elements 10a, each of which is an optical element, and a flange portion 10f that supports the plurality of lens elements 10a from the periphery. The flange portion 10 f includes an inter-lens flange portion 10 c between the lens elements 10 a and an outer peripheral flange portion 10 b at the outer peripheral portion of the first lens array 10. The plurality of lens elements 10a constituting the first lens array 10 are arranged such that their optical surfaces overlap on square or rectangular lattice points (16 × 4 × 4 in the illustrated example) arranged parallel to the XY plane. Are two-dimensionally arranged. Each lens element 10a has a convex optical surface 11a on the object-side main surface 10p, and a concave optical surface 11b on the image-side main surface 10q. Both optical surfaces 11a and 11b are, for example, aspherical surfaces. Yes. The flange portion 10f including the outer peripheral flange portion 10b and the inter-lens flange portion 10c is a flat plate-like flat portion. Here, the inter-lens flange portion 10c has a flange surface 11c on the object side and a flange surface 11d on the image side. A first spacer 30 is joined to the inter-lens flange portion 10c.

  The second lens array 20 on the image side shown in FIG. 1B is an integral body made of a thermoplastic resin, which is an optical material, and has a square or rectangular outline in plan view. The second lens array 20 includes a plurality of lens elements 20a, each of which is an optical element, and a flange portion 20f that supports the plurality of lens elements 20a from the periphery. The flange portion 20 f of the second lens array 20 includes an inter-lens flange portion 20 c between the lens elements 20 a and an outer peripheral flange portion 20 b at the outer peripheral portion of the second lens array 20. The plurality of lens elements 20a are two-dimensionally arranged on square or rectangular lattice points (16 × 4 × 4 in the illustrated example) arranged in parallel to the XY plane. Each lens element 20a has a concave optical surface 12a on the object-side main surface 20p, and a convex optical surface 12b on the image-side main surface 20q. Both optical surfaces 12a and 12b are, for example, aspherical surfaces. Yes. The flange portion 20f including the outer peripheral flange portion 20b and the inter-lens flange portion 20c is a flat plate-like flat portion. Here, the inter-lens flange portion 20c has a flange surface 12c on the object side and a flange surface 12d on the image side. The inter-lens flange portion 20 c is joined to the first and second spacers 30 and 40.

  The first spacer 30 is a member formed of glass, ceramics, metal, resin, or a composite material thereof. The first spacer 30 is bonded to the image side main surface 10q of the first lens array 10 on one main surface, and is bonded to the object side main surface 20p of the second lens array 20 on the other main surface. The first spacer 30 has a spacer body 31 and an opening 32. The spacer main body 31 is a lattice frame-shaped part, and the opening 32 has a rectangular shape. The opening 32 is formed as a through hole at a position corresponding to the lens element 10a and the like. The lens element 10a of the first lens array 10 and the lens element 20a of the second lens array 20 are supported by a rectangular frame around the opening 32 with respect to the central axis AX direction of the lens array parallel to the optical axis of each lens element. It will be.

  The second spacer 40 is a member formed of glass, ceramics, metal, resin, or a composite material thereof. The second spacer 40 is bonded to the image-side main surface 20q of the second lens array 20 on one main surface. The second spacer 40 has a spacer body 41 and an opening 42. The spacer main body 41 is a lattice frame-shaped part, and the opening 42 has a rectangular shape. The opening 42 is formed as a through hole at a position corresponding to the lens element 20a. The lens element 20a of the second lens array 20 is supported in the axis AX direction by a rectangular frame around the opening 42. Instead of providing separate spacers 30 and 40 as in the present embodiment, the resin that constitutes the lens element is a spacer that defines the distance between the lens array and the distance between the lens array and the image sensor around the lens element. The parts may be formed integrally.

  Each lens element 10a of the first lens array 10 and the lens element 20a of the second lens array 20 disposed to face the −Z side of the first lens array 10 sandwich the corresponding portion of the first spacer 30 therebetween. They are arranged along the axis AX and constitute one synthetic lens 1a. The plurality of synthetic lenses 1a arranged two-dimensionally on the lattice points corresponds to a single-eye lens of a field division method or a super-resolution method. Here, the field division method is a method in which one image is obtained by connecting images of different fields of view formed by the respective composite lenses 1a, which are individual composite optical systems, by image processing. is there. The super-resolution method is a method of obtaining one high-resolution image by image processing from images of the same field of view formed by the respective composite lenses 1a that are individual compound optical systems.

  As shown in FIGS. 1A and 1B and the like, the compound eye optical system 100 is incorporated in the imaging apparatus 1000. In addition to the compound-eye optical system 100 described above, the imaging apparatus 1000 includes an image sensor array 60 having a plurality of detection units (sensor elements) 61 provided corresponding to the individual synthetic lenses 1a constituting the compound-eye optical system 100. And an image processing unit 65 that performs processing on an image signal detected by the image sensor array 60. Here, the compound-eye optical system 100 is fixed to the image sensor array 60 and housed in a rectangular frame-shaped or square frame-shaped case 50.

  Hereinafter, the occurrence of optical crosstalk will be schematically described with reference to FIG. A periodic pattern described later is formed on the front and back optical surfaces of the first lens array 10. In the first lens array 10 of the compound-eye optical system 100, certain light L1 incident on the first lens element 10a1 reaches the optical surface 11b, is diffracted and reflected, and propagates while being totally reflected by the inter-lens flange portion 10c. It enters the adjacent second lens element 10a2. The light L1 is diffracted and transmitted from the inner side of the optical surface 11b of the second lens element 10a2 and emitted to the image side, passes through the lens element 20a of the second lens array 20, and passes through the second lens element of the sensor array 60. The light enters the detection unit (sensor element) 61 corresponding to 10a2. That is, in the compound eye optical system 100 of the present embodiment, the above-described crosstalk light is generated in the lens element 10a constituting any one or more of the composite lenses 1a and propagates through the inter-lens flange portion 10c. Then, the light enters the adjacent synthetic lens 1a and the detector 61 associated therewith.

  Here, when the light L1 incident from the optical surface 11a is reflected by the optical surface 11b of the first lens element 10a1, it is regularly reflected by the optical surface 11b and the period formed on the optical surface 11b. Diffracted by the target pattern PT1 exists. Further, when the light L1 is emitted from the optical surface 11b of the second lens element 10a2, it is refracted by the optical surface 11b and diffracted by the periodic pattern PT1 formed on the optical surface 11b. And exist.

  As another example, in the first lens array 10, a certain light L2 incident on the second lens element 10a2 reaches the optical surface 11b, is diffracted and reflected, and propagates while being totally reflected by the inter-lens flange portion 10c. It enters the adjacent third lens element 10a3. The light L3 is diffracted and reflected by the optical surface 11a of the third lens element 10a3, is emitted through the optical surface 11b, passes through the lens element 20a of the second lens array 20, and is a third lens in the sensor array 60. The light enters the detection unit (sensor element) 61 corresponding to the element 10a3. That is, in the compound eye optical system 100 of the present embodiment, the above-described crosstalk light is generated in the lens element 10a constituting any one or more of the composite lenses 1a and propagates through the inter-lens flange portion 10c. Then, the light enters the adjacent synthetic lens 1a and the detector 61 associated therewith.

  Here, when the light L2 is reflected by the optical surface 11b of the second lens element 10a2, the light L2 is diffracted by the regular reflection by the optical surface 11b and the periodic pattern PT1 formed on the optical surface 11b. There are things. Further, when the light L2 is reflected by the optical surface 11a of the third lens element 10a3, it is diffracted by the regular reflection by the optical surface 11a and the periodic pattern PT2 formed on the optical surface 11a. Things exist. When the light L2 is emitted from the optical surface 11b of the third lens element 10a3, the light L2 is refracted by the optical surface 11b and the light L2 is diffracted by the periodic pattern PT1 formed on the optical surface 11b. Exists.

  When the light L1 propagates between adjacent synthetic lenses 1a in the first lens array 10 to generate crosstalk, the light L1 is caused by diffraction on the optical surface 11b of a certain lens element 10a1. When the light L2 is diffracted and further diffracted by the optical surface 11b of the adjacent lens element 10a2, and the light L2 is diffracted by the optical surface 11b of a certain lens element 10a2, and by the optical surface 11a of the adjacent lens element 10a3. The case of diffracting is considered to be the main one.

FIG. 4 is a conceptual diagram for explaining the periodic pattern PT formed on the optical surfaces 11 a and 11 b of the first lens array 10. Periodic patterns PT are formed on the surfaces of the optical surfaces 11a and 11b as periodic streak irregularities having a pitch TP (μm) and a height H (μm). In addition, each groove | channel K which comprises the periodic pattern PT is extended uniformly in the direction perpendicular | vertical to a paper surface. Light having a wavelength λ (μm) incident on the optical surfaces 11a and 11b at an angle θ is emitted at an angle θ ′ by refraction. At this time, the emitted light includes ± first-order diffracted light, ± second-order diffracted light,. On this occasion,
n · sin θ−n ′ · sin θ ′ = mλ / TP (m is an integer representing the order)
Thus, diffraction occurs according to the pitch TP of the periodic unevenness (periodic pattern PT), and the smaller the pitch TP and the longer the wavelength λ, the larger the angle of diffraction and the easier the bending. Basically, diffraction occurs at an infinite order, but if the height H of the periodic unevenness (periodic pattern PT) is low, most of the diffracted light becomes 0th order light, and ± 1st order light and ± 2nd order light. The diffraction efficiency decreases greatly in the order of. Therefore, usually only the influence of the first order diffraction needs to be considered. For example, when the pitch TP of the periodic pattern PT is 2 μm and the wavelength λ is 0.5 μm, the primary light is diffracted at 14.5 degrees when the incident angle θ = 0 °. How much such primary light, secondary light, and the like are diffracted depends on the diffraction efficiency of the periodic pattern PT. However, in order to calculate the diffraction efficiency strictly, it is not a scalar theory but a vector. It is necessary to perform calculations based on theory.

  FIG. 5A is a conceptual diagram illustrating the positional relationship between the periodic patterns on the surfaces of a plurality of adjacent lens elements 10a. The first periodic patterns PT1 formed on the image-side main surface 10q of the first lens array 10 are formed at an equal pitch parallel to the Y direction. Further, the second periodic pattern PT2 formed on the object-side main surface 10p of the first lens array 10 is also formed at an equal pitch parallel to the Y direction, similarly to the image side.

  FIG. 6A is a diagram for conceptually explaining the relationship between the arrangement relationship of the periodic patterns on the surfaces of a plurality of adjacent lens elements 10a and the reduction of crosstalk. For example, in the first area AR1 of the first lens array 10, the first lens element 10a1 of the first lens array 10 is inclined with respect to the main surface from the direction orthogonal to the periodic patterns PT1 and PT2 from the object side. The incident light L (1) is diffracted and reflected by the optical surface 11b of the first lens element 10a1, and becomes the light L (2) subjected to the diffractive action. When the light L (2) propagates through the first lens array 10 and reaches one of the optical surfaces 11a and 11b of the adjacent second lens element 10a2, either of the optical surfaces 11a and 11b It becomes the light L (3) which received the diffraction effect again. The light L1 shown in FIG. 3 corresponds to the light L (3) when the light L (2) is diffracted and transmitted by the optical surface 11b. The light L2 shown in FIG. 3 corresponds to the light L (3) when the light L (2) is diffracted and reflected by the optical surface 11a. Here, the first periodic pattern PT2 of the optical surface 11b on which the light L (1) is incident is diffracted because it is perpendicular to the light L (1) when viewed from the optical axis direction (or light beam direction). Although the intensity of the light L (2) is lower than that of the light L (1), the light L (2) remains in the incident plane including the light L (2) and parallel to the optical axis. The same applies to the light L (3). According to the study by the present inventors, in the case of diffraction reflection, it has been found that the value of the diffraction efficiency is considerably smaller when the direction of the periodic pattern with respect to the incident light is perpendicular than when it is parallel. Accordingly, when light is incident perpendicular to the periodic patterns PT1 and PT2, it is difficult to diffuse the light from the incident plane to the outside. Since diffraction is repeated on a small optical surface, the light L (3) results in a sufficiently reduced intensity, and the light L1 and the light L2 in FIG. 3 deteriorate the accuracy of image detection or the like due to crosstalk. It is presumed that it will be of a level that is not reached.

  For example, in the second area AR2, when considering the light L (1) incident on the first lens element 10a1 of the first lens array 10 with an inclination from a direction parallel to the periodic patterns PT1 and PT2, The light L (1) is diffracted and reflected by the optical surface 11b of the first lens element 10a1, and becomes the light L (2) subjected to the diffractive action. When the light L (2) propagates through the first lens array 10 and reaches one of the optical surfaces 11a and 11b of the adjacent second lens element 10a2, it is caused by either of the optical surfaces 11a and 11b. , The light L (3) is again subjected to the diffraction action. The light L1 shown in FIG. 3 corresponds to the light L (3) when the light L (2) is diffracted and transmitted by the optical surface 11b. The light L2 shown in FIG. 3 corresponds to the light L (3) when the light L (2) is diffracted and reflected by the optical surface 11a. Here, the light L (1) incident in parallel to the second periodic pattern PT2 is diffused in a direction perpendicular to the optical axis. The light L (2) is further diffused in a direction perpendicular to the optical axis when diffracted and reflected by the optical surface 11a or diffracted and transmitted by the optical surface 11b. Therefore, the intensity of the light L (3) is sufficiently lowered, and it is presumed that the occurrence of crosstalk due to the light L1 and the light L2 in FIG. 3 is prevented.

  FIG. 5B is a diagram illustrating a reference example in which changes are made to the orientations of the periodic patterns PT1 and PT2. In this case, the second periodic pattern PT2 formed on the object-side main surface 10p of the first lens array 10 is formed at an equal pitch parallel to the Y direction on the optical surfaces arranged in a staggered pattern, and the remaining positions. Are formed at equal pitches parallel to the X direction. The first periodic pattern PT1 formed on the image-side main surface 10q of the first lens array 10 is parallel to the Y direction on the optical surface arranged at a staggered position corresponding to the object-side optical surface. They are formed at an equal pitch, and are formed at an equal pitch parallel to the X direction on the remaining optical surfaces.

  FIG. 6B corresponds to FIG. 6A, and is a diagram conceptually explaining the relationship between the periodic pattern arrangement relationship and the crosstalk reduction in the reference example. In the case of the reference example, for example, in the first region AR1 of the first lens array 10, the optical axis direction (in the first periodic pattern PT2 of the optical surface 11a with respect to the first lens element 10a1 of the first lens array 10 ( Or light L (1) incident perpendicularly as viewed from the light direction) is diffracted and reflected by the optical surface 11b of the first lens element 10a1 to become light L (2), and propagates through the first lens array 10; It reaches the optical surface 11a or the optical surface 11b of the adjacent second lens element 10a2. As described with reference to FIG. 6A, the light L (2) and the next light L (3) have lower intensity than the light L (1), but the light L (2) and the light L It remains in the incident plane including (3) and parallel to the optical axis. Here, in the case of diffraction reflection, since the value of the diffraction efficiency is considerably larger when the direction of the periodic pattern with respect to the incident light is perpendicular than when parallel, the crosstalk due to the light L2 in FIG. 3 in particular is sufficiently suppressed. It is speculated that it will not be possible.

  For example, in the second region AR2, the direction parallel to the second periodic pattern PT2 on the optical surface 11a with respect to the first lens element 10a1 of the first lens array 10 when viewed from the optical axis direction (or the light ray direction). When the light L (1) incident from the first lens element 10a1 is considered, the light L (2) diffracted and reflected by the optical surface 11b of the first lens element 10a1 propagates through the first lens array 10 and is adjacent to the second lens element 10b. The optical surface 11a or the optical surface 11b of 10a2 is reached. Here, the light L (2) diffracted and reflected by the first periodic pattern PT1 is diffused to some extent in the direction perpendicular to the optical axis, but the second pattern PT2 of the adjacent optical surface 11a or the second pattern PT2 of the optical surface 11b. When diffracted by one pattern PT1, it is presumed that the crosstalk is not sufficiently suppressed because it does not spread much compared to the case of the area AR2 in FIG. For this reason, the light L (3) does not have a sufficiently reduced intensity.

Regarding the pitch of the first and second periodic patterns PT1 and PT2 as TP (μm), when the use center wavelength in the plurality of lens elements 10a is λ (μm), the following relationship 1.5 <TP /Λ<8.0
Is satisfied. If the value TP / λ is smaller than 8.0, it can be avoided that the pitch TP becomes excessively rough, and it is relatively easy to improve the surface accuracy of the optical surfaces 11a and 11b. On the other hand, if the value TP / λ is larger than 1.5, it is possible to avoid the pitch TP from becoming excessively fine, and it is possible to avoid the occurrence of specifically very strong diffraction on the optical surfaces 11a and 11b.

Regarding the height of the first and second periodic patterns PT1 and PT2 to H (μm), when the use center wavelength in the plurality of lens elements 10a is λ (μm), the following relationship 0.5 < 100 × H / λ <7.0
Is satisfied. If the value 100 × H / λ is smaller than 7.0, it is possible to suppress the diffraction efficiency from becoming too high as a whole, and to prevent the crosstalk from increasing beyond the amount that can be reduced. If the value 100 × H / λ is larger than 0.5, it is relatively easy to leave the first and second tool marks TM1 and TM2 to some extent, and the diffraction efficiency is prevented from becoming too low. It is possible to ensure the effect of reducing crosstalk due to the complex diffraction of the periodic pattern.

Referring to FIG. 5A, when the periodic arrangement interval of the plurality of lens elements 10a is LP (mm) and the effective diameter of the plurality of lens elements 10a is ED (mm), <LP
Is satisfied. That is, a space is provided between effective surfaces of adjacent lens elements. The effective diameter ED is an average value of the effective diameters of the first and second optical surfaces 11a and 11b.

  For reference, a simulation was performed for diffraction efficiency when periodic patterns exist on the front and back of the lens elements of the lens array.

なお、入射角度は、例えばピッチＴＰ＝２μｍの場合、一般的な屈折率１．５の媒体（特に平板状のレンズ間フランジ部）内で全反射伝搬する際の反射角度が４２°であることを考慮して、５０°に設定した。表１から明らかなように、回折透過の場合は、垂直と平行の何れの場合も、ＴＥ波及びＴＭ波のどちらの偏光も、±１次の両方の次数において、回折効率は非常に小さいことがわかる。これに対して、回折反射の場合は、回折透過に比べて回折効率の値がかなり大きく、また、入射光を平行にした方が垂直にする場合よりも値が大きいことが分かる。 Table 1 below shows the diffractive efficiency due to polarized light when light is incident at a predetermined angle, by changing the incident direction (perpendicular or parallel) to the diffraction pattern and the pitch of the diffraction pattern, and calculating it strictly by vector theory ( The result calculated | required by RCWA method) is shown. Here, the height of the diffraction pattern, that is, the periodic pattern was assumed to be 10 nm. The periodic pattern is a diffraction grating having a rectangular cross section. The calculation wavelength was 0.55 μm. Table 1 shows a case where a light beam tilted with respect to the plane direction of the first lens array 10 from a direction perpendicular to and parallel to the tool mark is incident at an incident angle of 50 °. Shows the case. Actual diffraction occurs as ± 1st order, ± 3rd order, ± 5th order, etc. However, as a result of calculation, the diffraction efficiency of other orders is considerably smaller than the first order, so only the results of ± 1st order are obtained. Write.
[Table 1]

The incident angle is, for example, when the pitch TP = 2 μm, the reflection angle when the total reflection is propagated in a medium having a general refractive index of 1.5 (particularly, a plate-shaped flange between lenses) is 42 °. Was set to 50 °. As is clear from Table 1, in the case of diffractive transmission, the diffraction efficiency is very small in both ± 1st order both in the case of vertical and parallel polarizations of both TE wave and TM wave. I understand. On the other hand, in the case of diffraction reflection, the value of the diffraction efficiency is considerably larger than that of the diffraction transmission, and it is understood that the value obtained when the incident light is made parallel is greater than that when the incident light is made vertical.

  Note that the RCWA method used in the calculation of diffraction efficiency is the Rigorous Coupled Wave Analysis method, which is sometimes translated as a strict coupled wave analysis method. The RCWA method is not a scalar analysis but a kind of differential method in the electromagnetic field analysis considering that the electric field / magnetic field is a vector field. Although it can be applied only to a periodic structure, it can be used for rigorous calculation of diffraction efficiency and simulation of SPR (Surface Plasmon Resonance) phenomenon.

  The crosstalk caused by diffraction in only the first lens array 10 has been described above, but the same phenomenon occurs in the second lens array 20. For this reason, it is preferable to align the direction of the third periodic pattern PT3 formed on the image-side main surface 20q of the second lens array 20 in substantially the same direction. Moreover, it is desirable to align the direction of the 4th periodic pattern formed in the object side main surface 20p in substantially the same direction. At this time, as shown in FIGS. 7A and 7B, the directions of the third and fourth periodic patterns PT3 and PT4 of the second lens array 20 are set to the first and second directions of the first lens array 10, respectively. May coincide with the directions of the periodic patterns PT1, PT2, and as shown in FIGS. 8A and 8B, the third and fourth periodic patterns PT3, PT4 of the second lens array 20 may be used. May be orthogonal to the directions of the first and second periodic patterns PT1, PT2 of the first lens array 10. In the latter case, it is possible to prevent image deterioration caused by biasing in a specific direction in the synthetic lens 1a.

  Hereinafter, a method for manufacturing the first lens array 10 and the like will be described. The first lens array 10 is formed by injection molding.

  FIG. 9A is a view for explaining a mold for molding the first lens array 10. The mold apparatus 70 includes a first mold 71 and a second mold 72. The first mold 71 and the second mold 72 are mold-matched at the mold-matching surface PL, and a cavity 70 a is formed between the molds 71 and 72. A transfer surface 71c for transferring the shape of the first main surface 10p side of the first lens array 10 is formed on the first mold 71 so as to face the cavity 70a, and the second lens 72 has a second lens. A transfer surface 72c for transferring the shape on the second main surface 10q side of the array 10 is formed. The transfer surfaces 71c and 72c have a plurality of optical transfer portions 71g and 72g arranged two-dimensionally at a part thereof in order to transfer the optical surfaces 11a and 11b of the lens element 10a. In the mold apparatus 70, a gate GA communicating with the cavity 70a is formed. In this case, the gate GA is disposed not on the center of the transfer surfaces 71c and 72c but on the side, and injection molding is performed by a side gate method. An array lens having a large number of lens elements is required to have high accuracy required for each lens element, and the accuracy of decentering of each optical surface and the variation in the depth of the top of the optical surface is required to be, for example, 1 μm or less. It is done. For this reason, it is preferable to manufacture a metal mold | die as an integral type which is not nested. The integrated type improves the accuracy because the optical surface can be processed simultaneously. In consideration of avoiding an increase in cost for mold production, a mold in which a nesting and a receiving mold are combined as shown by a broken line in FIG. 9 may be used. In this case, a lens array obtained by molding is used. In order not to fall below the required accuracy, it is necessary to manufacture the mold with high accuracy and to accurately adjust the position of the nesting and the receiving mold.

  FIG. 9B is a cross-sectional view illustrating the entire structure of the mold apparatus 70. A runner RA is connected to the cavity 70a of FIG. 4A through a gate GA, and the runner RA is connected to a sprue SP on the resin supply side. As a result, the molten resin J from the sprue SP obtained by melting the thermoplastic resin fills the runner RA and fills the cavity 70a through the gate GA. By separating the first mold 71 and the second mold 72 after cooling the molten resin J, a sprue portion 81 corresponding to the sprue SP, a runner portion 82 corresponding to the runner RA, and a gate corresponding to the gate GA. A molded product 80 including a portion 83 and a lens array body 84 corresponding to the cavity 70a is formed. Here, the gate part 83 is subjected to a gate cut process, and the first lens array 10 is obtained by the resin injection part 14 that is the remaining part of the gate part 83 and the lens array body 84 ahead.

  The second lens array 20 is also molded by the same method as the first lens array 10. That is, the second lens array 20 is also manufactured by injection molding a thermoplastic resin by a side gate method.

  Hereinafter, an outline of a manufacturing process of the compound eye optical system 100 will be described with reference to FIG. First, initial molds to be the first mold 71 and the second mold 72 shown in FIG. 9A are manufactured (step S01). Specifically, the transfer surface 71c of the first mold 71 and the transfer surface 72c of the second mold 72 are created by a processing device that is a numerically controlled grinding or cutting device.

  FIG. 11 is a schematic perspective view illustrating the structure of a processing apparatus for processing the transfer surfaces 71c and 72c of the first and second molds 71 and 72 shown in FIG. The processing device 90 includes a cutting unit 91 for cutting a workpiece W that is a workpiece, and an NC drive mechanism 92 as a driving device that supports the cutting unit 91 with respect to the workpiece W. The cutting unit 91 includes a cutting tool 91a such as an end mill, for example, and can also rotate the cutting tool 91a around a rotation axis. The tip of the cutting tool 91a has a shape capable of machining a free curved surface. The NC drive mechanism 92 is a drive device having a structure in which a first stage 94b and a second stage 94c are placed on a pedestal 94a. Here, the first stage 94b supports the first movable portion 95a, and the first movable portion 95a supports the workpiece W via a chuck (not shown). The first stage 94b can move the workpiece W to a desired position along the z-axis direction, for example, at a desired speed. On the other hand, the second stage 94c supports the second movable portion 95b, and the second movable portion 95b supports the cutting unit 91. The second stage 94c supports the second movable portion 95b and the cutting unit 91, and can move them at a desired speed, for example, to a desired position along the x-axis direction. The second movable portion 95b is The cutting unit 91 can be supported and moved to a desired position along the y-axis direction at a desired speed.

  The first and second stages 94b and 94c and the first and second movable parts 95a and 95b are driven by driving a motor, a position sensor and the like built in the NC drive mechanism 92 by a drive control device (not shown). It is made to operate appropriately in the state of When processing a free-form surface on the surface Wa of the workpiece W, for example, cutting is performed while performing raster-type scanning. Specifically, main scanning can be performed by moving the cutting unit 91 in the x-axis direction at a constant speed by the second stage 94c, and the cutting unit 91 is moved at a constant speed in the z-axis direction by the first movable portion 95a. Sub-scanning can be performed. During the main scanning and the sub scanning, the second movable portion 95b appropriately displaces the cutting unit 91 in the y direction, so that a desired uneven shape can be formed on the surface Wa of the workpiece W. That is, the transfer surfaces 71c and 72c of the molds 71 and 72 shown in FIG. 9A can be processed as free-form surfaces.

  FIGS. 12A and 12B are views for explaining a scanning pattern of the tip of the cutting tool 91a by the processing apparatus 90 of FIG. As shown in the figure, after processing one line by scanning in the x direction, the processing position is shifted by one line in the z direction and returned to the starting point in the x direction, and then the next line is processed. Repeat the procedure. The surface Wa of the workpiece W is three-dimensionally processed by the cutting tool 91a of the cutting unit 91 that is two-dimensionally scanned, and as a result, the transfer surface 71c of the first mold 71 is obtained. A concave optical transfer portion 71g is formed on the transfer surface 71c. Note that the cutting of the workpiece W by the processing device 90 is not performed at a time, but is performed in a plurality of stages in the depth direction. For this reason, it is not necessary to process the surface Wa of the workpiece W as shown in FIG. 12 (A) or the like in the initial stage or in the middle stage, and the machining can be performed in a concentric or other locus. Linear processing as shown in FIG. 12A or the like is performed on the surface Wa of W. In addition to the method of repeating scanning in the same direction, the workpiece is processed in the direction opposite to the x direction after moving in the z direction (that is, the processing is performed while reciprocating in the x direction). Good.

  As shown in FIG. 12C, a groove K is formed on the optical transfer portion 71g of the transfer surface 71c corresponding to the locus of the tip of the cutting tool 91a, and extends in parallel in the x direction as a whole. Tool marks TM which are periodic grooves arranged at equal intervals in the direction are formed. In addition, after forming the transfer surfaces 71c and 72c by the processing apparatus 90, the tool marks can be erased by polishing other than the optical transfer portions 71g and 72g.

Returning to FIG. 10, the first lens array 10 and the like are molded by using the first mold 71 and the second mold 72 obtained in step S01 (step S02).
FIG. 13A shows the first lens array 10 as a molded product molded by the mold apparatus 70. On the optical surfaces 11a and 11b of the first lens array 10, a periodic streak pattern PT extending in parallel with the tool mark TM of FIG. 12C transferred is formed (see FIG. 12D). . In this case, in the first lens array 10, a resin injection portion 14 corresponding to the protruding gate trace is formed at the center of one of the four sides. The protrusions of the resin injection part 14 can be removed by post-processing. The first lens array 10 has a difference in transferability between the gate-side region R1 and the counter-gate-side region R2, and an asymmetric shape error occurs with respect to each optical surface 11a on the front side of the first lens array 10. To do. Similarly, although not shown, an asymmetrical shape error occurs with respect to each optical surface 11b on the back side.
FIG. 13B shows another example of molding the first lens array 10. In this case, the resin injecting portion 14 is formed at a position different from the first lens array 10 of FIG. 13 by 90 °. In this molding example, the orientation of the periodic patterns PT1 and PT2 is in line with the flow of the resin at the time of resin injection, which is advantageous in reducing molding defects. 13A and 13B, after molding is completed and the lens array is taken out from the mold and the resin injection portion 14 is cut from the runner portion, a part of the resin injection portion 14 or the resin injection portion 14 is connected. Since the trace of the part which was done is left in a lens array, the direction of a periodic pattern and the tendency of a shape error can be grasped | ascertained by this part and trace.

  Next, the shape errors of the optical surfaces 11a and 11b in the first lens array 10 as the obtained molded product are measured and evaluated. Using this result, the corrected shape of the mold is calculated (step S03). That is, since the thermoplastic resin is injection-molded by the side gate method, an asymmetric shape error occurs in the first lens array 10, but the cavity 70a having a compensation shape that cancels out such a shape error in advance is provided. If it is formed in advance by the mold apparatus 70, the first lens array 10 having no shape error can be obtained. A correction shape corresponding to such a compensation shape is a non-axisymmetric shape.

  Next, based on the correction shape data obtained in step S02, correction processing including non-axisymmetric errors and the like is performed on the transfer surfaces 71c and 72c of the molds 71 and 72 (step S04). This correction processing is performed by raster processing. That is, the entire surface of the transfer surfaces 71c and 72c is processed by the scanning pattern of the cutting tool 91a shown in FIG. 12A (scanning the cutting tool 91a in the main direction and feeding it in the sub direction). As described above, processing in the same direction may be performed not only by changing the position in the Z direction, but also by reciprocating in the X direction while changing the position in the Z direction.

  Next, the first lens array 10 and the like are molded using the molds 71 and 72 obtained in step S02 (step S05). As described above, there is a difference in transferability between the gate-side region R1 and the counter-gate-side region R2, but since the mold is corrected by feedback in step S04 in advance, the shape errors of the optical surfaces 11a and 11b are reduced. Approach the design shape.

  Next, the shape errors (including the arrangement relationship) of the optical surfaces 11a and 11b in the first lens array 10 as the obtained molded product are measured and evaluated, and are within an allowable range with respect to the designed lens shape. Whether or not (step S05). When the shape errors of the optical surfaces 11a and 11b are within the allowable range, the shapes of the transfer surfaces 71c and 72c of the mold apparatus 70 are determined, and the molding conditions are also determined. On the other hand, when the shape errors of the optical surfaces 11a and 11b are outside the allowable range, the process returns to step S04 and the transfer surfaces 71c and 72c of the molds 71 and 72 are subjected to correction processing again.

  When the shape errors of the optical surfaces 11a and 11b are within the allowable range in step S05, the obtained first lens array 10 is sent to the assembly process of the compound-eye optical system 100 (step S06). In the assembly process, the first lens array 10 obtained as described above and the second lens array 20 obtained by the same method are joined with the first spacer 30 interposed therebetween, and further, on the second lens array 20 side. The second spacer 40 is joined. In addition, once it is confirmed that the shape error is within the allowable range, the subsequent measurement may be repeated many times by reducing or omitting the frequency.

  According to the compound eye optical system 100 and the like of the first embodiment described above, in the first lens array 10, the first periodic pattern PT1 on the optical surface 11b side faces substantially the same direction, so the first period The combined diffraction at the target pattern PT1 can reduce stray light that causes crosstalk, and improve the performance of the compound-eye optical system 100. Further, since the first pattern PT1 and the second pattern PT2 are directed in substantially the same direction, the stray light that causes crosstalk can be reduced by the combined diffraction of the first pattern PT1 and the second pattern PT2. And the performance of the compound-eye optical system 100 can be improved.

[Second Embodiment]
The compound eye optical system according to the second embodiment of the present invention will be described below. The compound eye optical system according to the second embodiment is a partial modification of the compound eye optical system according to the first embodiment, and parts not specifically described are the same as those of the first embodiment.

  FIG. 14A is a plan view illustrating the first lens array 10 constituting the compound-eye optical system of the second embodiment, and FIG. 14B illustrates the first lens array 10 of the present embodiment. It is a back view. As shown in the figure, a periodic pattern PT2 is formed not only on the optical surface 11a but also on the entire surface of the object-side main surface 10p of the first lens array 10. Similarly, a periodic pattern PT1 is formed on the image-side main surface 10q of the first lens array 10 not only on the optical surface 11b but also on the entire surface. In other words, in the first lens array 10 of this example, the fifth periodic pattern is formed in a region other than the optical surface.

  Although not shown, periodic patterns PT3 and PT4 can be formed on the entire front and back main surfaces of the second lens array 20 as in the first lens array 10 described above. At this time, the periodic pattern of the second lens array 20 can be parallel to the first lens array 10, but the periodic pattern of the second lens array 20 can also be orthogonal to the first lens array 10.

  In the case of the present embodiment, by using a mold provided with a periodic pattern over the entire mold forming surface, it is possible to save the trouble of erasing a tool mark by polishing portions other than the optical transfer surface. In addition, since a periodic pattern is formed in addition to the optical surface in the lens arrays 10 and 20 obtained by molding using such a mold, the light propagating through the lens arrays 10 and 20 is also formed. The effect of further reducing the strength can be expected. A periodic pattern (fifth periodic pattern) in a region other than the optical surface may be provided only on one of the front and back main surfaces of the first and second lens arrays 10 and 20. In this case, it is desirable to provide at least on the image side main surfaces 10q and 20q.

The compound eye optical system according to the embodiment has been described above, but the compound eye optical system according to the present invention is not limited to the above. For example, the arrangement of the lens elements 10a and 20a constituting the lens arrays 10 and 20 and the shape of the optical surfaces 11a, 11b, 12a and 12b should be changed as appropriate according to the use or specification of the compound-eye optical system 100 or the imaging apparatus 1000. Can do. For example, the lens elements 10a and 20a are not limited to being arranged at 4 × 4 lattice points, but can be arranged at lattice points of 3 × 3, 5 × 5, and the like, and are not limited to rectangular lattices. It can be. Furthermore, the lens array constituting the compound-eye optical system 100 is not limited to two layers (first and second lens arrays 10 and 20), and may be one layer (only the first lens array 10) or three layers. It can also be set as above.

  The opening 32 of the first spacer 30 and the opening 42 of the second spacer 40 can have various shapes such as a square, a rectangle, and a circle. Specifically, the openings 32 and 42 can be set to be elliptical or the like.

  In the above embodiment, the periodic pattern on the optical surface of each lens element of the lens array is formed by transferring the tool mark provided on the optical transfer surface of the mold, but the present invention is not necessarily limited thereto. After forming the lens array with a mold in which the tool marks on the optical transfer surface have been erased by other methods, the streaks are formed on the optical surface of each lens element of the lens array by laser irradiation, etching, or mechanical processing. You may make it form the periodic pattern which consists of this unevenness | corrugation.

DESCRIPTION OF SYMBOLS 1a ... Synthetic lens, 10 ... 1st lens array, 20 ... 2nd lens array, 10a, 20a ... Lens element, 10b, 20b ... Outer peripheral flange part, 10c, 20c ... Inter-lens flange part, 10p, 20p ... Main object side Surface, 10q, 20q ... image side main surface, 11a, 11b, 12a, 12b ... optical surface, 11c, 12c, 11d, 12d ... flange surface, 14 ... resin injection part, 30,40 ... spacer, 32,42 ... opening 50 ... Case, 60 ... Sensor array, 61 ... Detection unit, 65 ... Image processing unit, 70 ... Mold device, 71,72 ... Mold, 71c, 72c ... Transfer surface, 71g, 72g ... Optical transfer unit, 80 ... Molded product, 90 ... Processing device, 91 ... Cutting unit, 91a ... Cutting tool, 92 ... Drive mechanism, 94b, 94c ... Stage, 95a 95b ... movable part, 100 ... compound optical system, 1000 ... imaging device, AX ... axis, K ... groove, L1, L2, L3 ... light, R1 ... gate side region, R2 ... anti-gate side region, TM ... tool mark, PT1, PT2, PT3, PT4 ... periodic pattern, W ... work, Wa ... surface

Claims (18)

A lens array comprising an optical material including a plurality of lens elements arranged two-dimensionally and a flange portion extending between the plurality of lens elements;
On the image side main surface of the two main surfaces of the lens array, a plurality of first optical surfaces of the plurality of lens elements is formed with a first pattern composed of periodic irregularities extending linearly,
The compound eye optical system, wherein the first patterns of the plurality of first optical surfaces face substantially the same direction.   2. The compound eye according to claim 1, wherein the lens array is molded by injecting a thermoplastic resin into a molding die, and a resin injection portion is provided on a side portion of an outer edge connecting the two main surfaces. Optical system. When the pitch of the unevenness of the first pattern is TP (μm) and the use center wavelength in the plurality of lens elements is λ (μm), the following relationship 1.5 <TP / λ <8.0
The compound eye optical system according to claim 1, wherein the compound eye optical system is satisfied. When the height of the unevenness of the first pattern is H (μm) and the use center wavelength of the plurality of lens elements is λ (μm), the following relationship 0.5 <100 × H / λ <7. 0
The compound eye optical system according to any one of claims 1 to 3, wherein:   The compound eye optical system according to any one of claims 1 to 4, wherein the plurality of lens elements are arranged in a matrix along four orthogonal reference directions.   On the object-side main surface of the lens array, a plurality of second optical surfaces of the plurality of lens elements are formed with second patterns made of periodic irregularities extending linearly, and the plurality of second optical surfaces. The compound eye optical system according to any one of claims 1 to 5, wherein the second pattern is oriented in substantially the same direction as the first pattern.   The compound eye optical system according to claim 1, wherein a plurality of lens arrays including the lens array are stacked. The plurality of lens arrays includes a first lens array having the lens element of the lens array as a first lens element and the flange portion as a first flange portion, and a plurality of second lenses arranged in a two-dimensional manner. A second lens array that is an integral part of an optical material including an element and a second flange portion extending between the plurality of second lens elements;
A third pattern which is a periodic unevenness extending linearly to a plurality of third optical surfaces corresponding to the plurality of second lens elements on at least an image side main surface of the two main surfaces of the second lens array. The compound eye optical system according to claim 7, wherein the third patterns of the plurality of third optical surfaces face substantially the same direction.   On the object-side main surface of the second lens array, a plurality of fourth optical surfaces corresponding to the plurality of second lens elements are formed with a fourth pattern that is a periodic unevenness extending linearly, 9. The compound-eye optical system according to claim 8, wherein the fourth pattern of the fourth optical surface faces substantially the same direction as the third pattern.   10. The compound eye optical system according to claim 9, wherein the first and second patterns of the first lens array and the third and fourth patterns of the second lens array are directed in substantially the same direction.   The compound-eye optical system according to claim 9, wherein the first and second patterns of the first lens array and the third and fourth patterns of the second lens array face different directions.   The fifth pattern comprising periodic irregularities extending linearly on the surface of the flange portion is formed on at least one of the two main surfaces. The compound eye optical system described.   At least one of the two main surfaces of the first lens array and the two main surfaces of the second lens array has a fifth pattern formed of periodic irregularities extending linearly on the surface of the flange portion. The compound eye optical system according to any one of claims 8 to 11, wherein the compound eye optical system is provided. A compound eye optical system according to any one of claims 1 to 13,
An image pickup apparatus comprising: a sensor array having a plurality of sensor elements provided corresponding to the plurality of lens elements. Filling a molding space of a molding die having a molding surface corresponding to a plurality of lens elements and a flange portion extending between the plurality of lens elements with a lens material;
Among the two principal surfaces, the first principal surface on the image side includes a plurality of first patterns each including a first pattern that is included in the plurality of lens elements and that has a periodic unevenness extending in a straight line facing substantially the same direction. Obtaining a lens array having an optical surface, and a method of manufacturing a compound eye optical system.   The molding surface has a tool mark made of periodic irregularities extending linearly at a portion corresponding to the plurality of first optical surfaces, and after the shape of the tool mark is transferred to a lens material, the molding die The manufacturing method of the compound eye optical system of Claim 15 which obtains a lens array provided with the said several lens element integrally molded and the said flange part by taking out a molded object from A.   The method of manufacturing a compound eye optical system according to claim 16, wherein the tool mark is formed on a molding surface of the mold by raster processing.   18. The method according to claim 15, further comprising a step of machining a molding surface of the mold, wherein the machining step includes a machining data correction step of correcting a non-axisymmetric error by feedback. The manufacturing method of the compound-eye optical system as described in 1 above.

Images