JP6134466B2 - Stereoscopic image display device, stereoscopic image display optical element - Google Patents

Stereoscopic image display device, stereoscopic image display optical element Download PDF

Info

Publication number
JP6134466B2
JP6134466B2 JP2013054468A JP2013054468A JP6134466B2 JP 6134466 B2 JP6134466 B2 JP 6134466B2 JP 2013054468 A JP2013054468 A JP 2013054468A JP 2013054468 A JP2013054468 A JP 2013054468A JP 6134466 B2 JP6134466 B2 JP 6134466B2
Authority
JP
Japan
Prior art keywords
image display
cylindrical lens
length
stereoscopic image
predetermined
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
JP2013054468A
Other languages
Japanese (ja)
Other versions
JP2014182160A (en
Inventor
武 船橋
武 船橋
優二 池田
優二 池田
Original Assignee
グローバルウェーブ株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by グローバルウェーブ株式会社 filed Critical グローバルウェーブ株式会社
Priority to JP2013054468A priority Critical patent/JP6134466B2/en
Publication of JP2014182160A publication Critical patent/JP2014182160A/en
Application granted granted Critical
Publication of JP6134466B2 publication Critical patent/JP6134466B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Description

  The present invention relates to a stereoscopic image display device, a stereoscopic image display optical element, and a stereoscopic image display optical element manufacturing mold.

  In recent years, flat image display devices using a liquid crystal display (LCD) and an organic EL display (Organic Electro-Luminescence Display: OELD) have become widespread. Such a flat image display device can be reduced in power consumption and thinned, so that it is not only an image display device for business use, a television receiver for home use, but also a notebook personal computer, a high-performance mobile phone, Widely used in multifunctional portable information terminals and the like. FIG. 21 is a diagram showing an image display surface of a conventionally used flat image display device. As shown in FIG. 21, subpixels correspond to the respective colors of R (red), G (green), and B (blue). A technique for adjusting the luminance for each sub-pixel and displaying a color image with a set of sub-pixels R (red), G (green), and B (blue) as pixels is a well-known technique. Black portions between the R (red), G (green), and B (blue) sub-pixels are portions that do not emit light, and are referred to as light shielding portions or black matrices.

A stereoscopic image display device called a lenticular lens system using a lenticular lens (a plurality of cylindrical lenses) is known (see Patent Document 1 and Patent Document 2). In a display device that provides different images toward a plurality of viewpoints using an optical separating unit such as a lenticular lens, a three-day moire (3D moire (3D
A phenomenon called “moire)) occurs. 3D moiré is periodic luminance unevenness (sometimes referred to as color unevenness) caused by displaying different images in different angular directions. 3D moire is a variation in the angular direction of luminance, and if the variation in the angular direction of luminance is large, there is an undesirable effect on stereoscopic vision (see Patent Document 3). In order to improve such a problem caused by the optical separation means, a display device has been proposed in which the shape and arrangement of the pixel electrode and the light-shielding portion of the display portion are devised to suppress the deterioration in display quality ( For example, see Patent Document 4). In addition, a method of manufacturing a lenticular lens using a mold is known (see, for example, Patent Document 5).

Japanese Patent Laid-Open No. 7-5420 Japanese Patent Laid-Open No. 9-236777 JP 2012-18382 A JP-A-2005-208567 JP 2008-44136 A

  In the technique described in Patent Document 4, a general liquid crystal display as shown in FIG. 21 that has been widely used in the past in order to reduce 3D moire by using a pixel electrode (subpixel) of a display unit as a special shape. The stereoscopic image display device cannot be configured by adopting the flat image display device or the organic EL display flat image display device. For this reason, the price of a stereoscopic image display device having such a specially shaped sub-pixel structure is also high. In addition, a stereoscopic image display device could not be provided unless a flat image display device in which the pixel electrode has a special shape according to the size of various screens was newly developed.

  The present invention has been made in view of these points. That is, a stereoscopic image display device and a stereoscopic image display optical element for reducing 3D moire in a stereoscopic image display device using a flat image display device using various sizes of liquid crystal displays and organic EL displays currently on the market are provided. Then, a manufacturing mold for manufacturing such a stereoscopic image display optical element is provided.

  The stereoscopic image display device of the present invention is a stereoscopic image display device formed by sequentially arranging a planar image display device in which subpixels are regularly arranged on an image display surface and a cylindrical lens having a curvature in a predetermined axis direction in the predetermined axis direction. An image display optical element, and each of the cylindrical lenses has a length in the direction having the curvature, or a distance between optical axes between adjacent cylindrical lenses varies around a predetermined reference length, and a predetermined number The maximum value of the difference between the position of the boundary of each cylindrical lens and the position of the boundary of the subpixel is 1.4% or more of the length of the subpixel in the predetermined axis direction, 25 The range is less than%.

  The stereoscopic image display optical element of the present invention is a stereoscopic image display optical element mounted on a flat image display device in which sub-pixels are regularly arranged on an image display surface, and each cylindrical lens having a curvature in a predetermined axial direction Sequentially arranged in a predetermined axial direction, the length of each cylindrical lens in the direction having the curvature, or the distance between optical axes between adjacent cylindrical lenses varies around a predetermined reference length, and the predetermined reference The length is equal to a predetermined number of the sub-pixels, and the maximum value of the variation is (2.8% / predetermined number) or more of the predetermined reference length (50% / predetermined number). ).

  The stereoscopic image display optical element manufacturing mold of the present invention is a stereoscopic image display optical element manufacturing mold for manufacturing a stereoscopic image display optical element to be attached to a flat image display device in which subpixels are regularly arranged on an image display surface. Then, each cylindrical lens having a curvature in a predetermined axial direction is sequentially arranged in the predetermined axial direction, and the length of each cylindrical lens in the direction having the curvature, or the distance between optical axes between adjacent cylindrical lenses is The predetermined reference length has a variation around the predetermined reference length, and the predetermined reference length is equal to a predetermined number of the sub-pixels, and the maximum value of the variation is (2.8% of the predetermined reference length). / The predetermined number) or more and less than (50% / the predetermined number).

  According to the technology of the present invention, the length of each cylindrical lens in the direction having the curvature or the distance between optical axes between adjacent cylindrical lenses is varied to reduce the 3D moire, and the stereoscopic image display device and the stereoscopic image display An optical element and a stereoscopic image display optical element manufacturing mold can be provided.

It is a figure showing the image display surface of the three-dimensional image display apparatus of embodiment. It is a principle figure which shows the aspect of each sub pixel of the three-dimensional image display apparatus of embodiment. It is a principle figure which sees the arrangement | positioning relationship between the image display surface of the three-dimensional image display apparatus in embodiment, and the three-dimensional image display optical element which has a some cylindrical lens from an image display surface direction. It is sectional drawing in the step in which a fundamental stereo image display optical element is provided to a market as an optical component. FIG. 3 is a principle view of an arrangement relationship between an image display surface of a stereoscopic image display device and a stereoscopic image display optical element having a plurality of cylindrical lenses according to an embodiment from a cross-sectional direction orthogonal to the image display surface. It is a figure which sees the arrangement | positioning relationship between the image display surface of the stereo image display apparatus in 1st Embodiment, and the stereo image display optical element which has several cylindrical lenses from an image display surface direction. It is a figure which sees the arrangement | positioning relationship between the image display surface of the stereo image display apparatus in 1st Embodiment, and the stereo image display optical element which has several cylindrical lenses from the cross-sectional direction orthogonal to an image display surface. It is a figure which sees the arrangement | positioning relationship between the image display surface of the stereo image display apparatus in 2nd Embodiment, and the stereo image display optical element which has a some cylindrical lens from an image display surface direction. It is a figure which sees the arrangement | positioning relationship between the image display surface of the stereo image display apparatus in 2nd Embodiment, and the stereo image display optical element which has several cylindrical lenses from the cross-sectional direction orthogonal to an image display surface. It is a figure which shows the stereo image display optical element which inclines and arranges a cylindrical lens in the direction which makes angle (theta) with respect to the Y-axis. It is a figure which shows the three-dimensional image display optical element which shifts the position of a cylindrical lens not only to an X-axis direction but the Y-axis direction. It is a figure which shows another three-dimensional image display optical element which shifts the position of a cylindrical lens not only to an X-axis direction but the Y-axis direction. It is a figure which shows the three-dimensional image display optical element which the most concave part of a cylindrical lens swells smoothly. It is a figure which shows the three-dimensional image display optical element which inclines the image display surface 90 degree in 5th Embodiment, and faces two sub pixels of the Y-axis direction in which each cylindrical lens has a curvature. It is a figure of the cross-sectional direction orthogonal to an image display surface about the arrangement | positioning relationship between the image display surface of the three-dimensional image display apparatus in 5th Embodiment, and a three-dimensional image display optical element. It is a figure which shows a part of actual measurement data which show the deviation from the average of the distance between the optical axes of the cylindrical lens of the stereoscopic image display optical element of the Example in 5th Embodiment. It is a figure which shows a part of measured data which shows the deviation from the average of the distance between the optical axes of the cylindrical lens of the stereoscopic image display optical element of the comparative example in 5th Embodiment. It is a figure which compares 3D moire of the Example in 5th Embodiment, and a comparative example. It is a figure which calculates | requires and contrasts the spatial frequency of the stereoscopic image display optical element of the Example of 5th Embodiment from the calculation, and the spatial frequency of the stereoscopic image display optical element of a comparative example. It is a block diagram which shows the whole stereoscopic image display apparatus of embodiment. It is a figure which shows the image display surface of the planar image display apparatus used conventionally.

  The stereoscopic image display device according to the embodiment is a stereoscopic image formed by sequentially arranging a planar image display device in which sub-pixels are regularly arranged on an image display surface and a cylindrical lens having a curvature in a predetermined axis direction in the predetermined axis direction. A display optical element. Each of the cylindrical lenses faces a predetermined number of subpixels having a length in the direction of curvature, or a distance between optical axes between adjacent cylindrical lenses that varies with a predetermined reference length as a center. The maximum value of the difference between the boundary position and the subpixel boundary position is in a range of 1.4% or more and less than 25% of the length of the subpixel in the predetermined axis direction. The predetermined reference length is a length in a predetermined axial direction which is an axial direction having a curvature of each cylindrical lens when each cylindrical lens is uniform and has no variation, and is equal to a predetermined number of subpixels. .

  The stereoscopic image display optical element according to the embodiment is a stereoscopic image display optical element that is attached to a flat image display device in which subpixels are regularly arranged on the image display surface. In this stereoscopic image display optical element, each cylindrical lens having a curvature in a predetermined axial direction is sequentially arranged in the predetermined axial direction, and the length of each cylindrical lens in the direction having the curvature, or the optical axis between adjacent cylindrical lenses. The inter-space distance varies around a predetermined reference length, the predetermined reference length is equal to a predetermined number of subpixels, and the maximum value of the variation is (2.8% /% of the predetermined reference length). The predetermined number) or more and less than (25% / predetermined number).

  The stereoscopic image display optical element manufacturing mold of the embodiment is a stereoscopic image display optical element manufacturing mold for manufacturing a stereoscopic image display optical element to be mounted on a flat image display device in which subpixels are regularly arranged on the image display surface. . In this stereoscopic image display optical element manufacturing mold, each cylindrical lens having a curvature in a predetermined axial direction is sequentially arranged in the predetermined axial direction, and the length of each cylindrical lens in the direction having the curvature, or between adjacent cylindrical lenses. The optical axis distance has a variation around a predetermined reference length, the predetermined reference length is equal to a predetermined number of the sub-pixels, and the maximum value of the variation is (2 .8% / predetermined number) or more and less than (25% / predetermined number).

  The embodiment will be described below with reference to the drawings, focusing on the description of the stereoscopic image display optical element that is a main part of the embodiment.

  The dimensions of each part shown in each drawing and the direction of the arrow indicating the direction of the light beam are emphasized for easy understanding of the explanation of the drawing, and the actual stereoscopic image display device and the stereoscopic image display optical element It does not correspond to the dimensions of each part and the actual direction of light rays.

(About the principle of stereoscopic image display device)
The principle of the stereoscopic image display apparatus according to the embodiment will be described with reference to FIGS.

  FIG. 1 is a diagram illustrating an image display surface of the stereoscopic image display apparatus according to the embodiment.

  Each position of the left eye and the right eye of the viewer who visually recognizes the image of the stereoscopic image display device of the embodiment is on the surface side of the paper surface of FIG. The enlarged explanation part (the range of the rectangle in the circle indicated by the arrow) shown in FIG. 1 is the part of the image display surface 10 of the stereoscopic image display device where the range of the part shown in FIGS. It shows whether there is. 2, 3, and 5, the arrangement relationship between the enlarged explanation portion of the image display surface 10 and a plurality of cylindrical lenses (described later) arranged in the portion is schematically illustrated as a partial view. In other drawings, only the enlarged explanation part is described.

  J sub-pixels are arranged in the X-axis direction of the image display surface 10, and K sub-pixels are arranged in the Y-axis direction of the image display surface 10 (see FIG. 21). That is, the total number of subpixels on the image display surface 10 is J × K. Each of the arrays of J subpixels in the X axis direction is referred to as a row, and each of the arrays of K subpixels in the Y axis direction is referred to as a column.

  In the flat image display device, the sub-pixels are arranged on the image display surface 10 as follows, for example. The width of three subpixels in the X-axis direction (the length of the short side of the subpixel shown in FIG. 2) and the length of one subpixel in the Y-axis direction (the long side of the subpixel shown in FIG. 2) Is equal to the length). That is, the subpixel is rectangular, and the length of the short side of the subpixel is 1/3 of the length of the long side of the subpixel. For example, J (1024 × 3) subpixels are arranged in the X-axis direction, and K (768) subpixels are arranged in the Y-axis direction. In a normal flat image display device, since one pixel is formed by three colors of R (red), G (green), and B (blue) in the X-axis direction, J / 3 ( 1024) pixels are arranged (see FIG. 21).

  The origin of the X axis and the origin of the Y axis are the upper left corner of the image display surface 10 shown in FIG. The first to Jth numbers are used as subpixel numbers in the order of increasing separation distance from the X-axis origin in the X-axis direction, and are used in the following description. In order of increasing separation distance from the Y-axis origin to the Y-axis direction. The first to Kth numbers are used as subpixel numbers in the following description.

  FIG. 2 is a principle diagram illustrating an aspect of each subpixel of the stereoscopic image display apparatus according to the embodiment.

  In a conventional flat image display device, one pixel is formed with continuous R, G, and B as one pixel. However, in a stereoscopic image display device in which a stereoscopic image display optical element having a plurality of cylindrical lenses and a planar image display device are combined, the configuration of one pixel recognized by a viewer is different. That is, the pixels are arranged so that a pixel corresponding to the right eye (right eye pixel) and a pixel corresponding to the left eye (left eye pixel) can be separated and viewed. As for the arrangement of the right-eye pixel and the left-eye pixel, arrangement methods such as a two-parallax method, a four-parallax method, a five-parallax method, and a seven-parallax method are conventionally known. In the following description, the 5-parallax method will be described as a specific example, but the content of the description does not lose generality regardless of the number of parallaxes.

  In order to clarify the difference between the pixel configuration that the viewer recognizes as one pixel in the stereoscopic image display device of the embodiment and the pixel configuration in the conventional flat image display device, first, the pixel configuration in the conventional flat image display device Will be explained.

  In the conventional flat image display apparatus, each pixel (picture element) is configured as shown in FIG. The first sub-pixel R (red) in the X-axis direction, the second sub-pixel G (green) in the X-axis direction, and the third sub-pixel B (blue) in the X-axis direction constitute one pixel (pixel). The Similarly, each pixel is configured on the entire image display surface 10. Arranging pixels in this way is a well-known technique. Also, a technique for controlling the hue, saturation, and brightness of each pixel by controlling the luminance of light emitted from each sub-pixel constituting each pixel and displaying a desired image as a flat image on the entire image display surface 10 is also well known. Technology.

  In the stereoscopic image display device according to the embodiment, unlike the pixel configuration shown in FIG. 21, as shown in FIG. 2, the first sub-pixel R (red) in the X-axis direction denoted by reference numeral 1 and X denoted by reference numeral 1 The viewer can visually recognize the sixth sub-pixel B (blue) in the axial direction and the eleventh sub-pixel G (green) in the X-axis direction denoted by reference numeral 1 as one pixel. The reason why the numbers 1 to 5 are repeatedly given to the subpixels in FIG. 2 is to facilitate explanation in the 5-parallax method.

  The viewer also has a second sub-pixel G in the X-axis direction labeled 2, a seventh sub-pixel R in the X-axis direction labeled 2, and a 12th sub-pixel B in the X-axis direction labeled 2. It is made visible as 1 pixel.

  The viewer also has a third sub-pixel B in the X-axis direction labeled 3, an eighth sub-pixel G in the X-axis direction labeled 3, and a 13th sub-pixel R in the X-axis direction labeled 3. It is made visible as 1 pixel.

  In addition, the viewer sees the fourth sub-pixel R in the X-axis direction labeled 4, the ninth sub-pixel B in the X-axis direction labeled 4, and the 14th sub-pixel G in the X-axis direction labeled 4. It is made visible as 1 pixel.

  Further, the viewer sees the fifth sub-pixel G in the X-axis direction denoted by reference numeral 5, the tenth sub-pixel R in the X-axis direction denoted by reference numeral 5, and the fifteenth sub-pixel B in the X-axis direction denoted by reference numeral 5. It is made visible as 1 pixel.

  In the same manner, the viewer can visually recognize the latest three colors of R, G, and B having the same code as one pixel. Here, the order of arrangement of the three types of colors R, G, and B is not essential, and may be aligned with R, B, and G, or may be aligned with B, R, and G. , B, G, R may be arranged, G, R, B may be arranged, or G, B, R may be arranged.

The sub-pixel length P SPX in the X-axis direction, which is the length of one sub-pixel in the X-axis direction, is the same for all sub-pixels, and is the length in the Y-axis direction that is the length of one sub-pixel in the Y-axis direction. The subpixel length P SPY is equal for all subpixels.

The lengths (5 subpixel lengths) P P1 of the 5th subpixel from the first to the 5th, the 5 subpixel lengths P P2 from the 6th to the 10th, and the 5 subs from the 11th to the 15th The pixel length P P3 and the five subpixel lengths P P4 from the 26th to the 30th are all equal. In general terms, the five subpixel lengths P Pn are all equal.

The length of the light shielding portion in the X-axis direction of the subpixel (light shielding portion length) PSX is the same in all the subpixels. The length of the light shielding part in the Y-axis direction of the subpixel (light shielding part length) PSY is the same in all the subpixels. Therefore, the length of the light-shielding portion of two adjacent subpixels in the X-axis direction is 2 × P SX . In addition, the length of the light-shielding portion of two adjacent subpixels in the Y-axis direction is 2 × PSY . The light blocking portion length P SX is included in the sub pixel length P SPX in the X-axis direction of the sub pixel. The light blocking portion length PSY is included in the subpixel length PSPY in the Y-axis direction of the subpixel.

  FIG. 3 is a principle view of the positional relationship between the image display surface of the stereoscopic image display device and the stereoscopic image display optical element having a plurality of cylindrical lenses in the embodiment viewed from the image display surface direction. A stereoscopic image display optical element having a plurality of cylindrical lenses is also referred to as a lenticular lens.

  In FIG. 3, in order to make the drawing easier to see, the description of the light shielding portion between the sub-pixels is omitted, and the boundary of each cylindrical lens is emphasized and written boldly. In the stereoscopic image display optical element 11, only the enlarged display portion is displayed, but the stereoscopic image display optical element 11 is disposed so as to face the entire surface of the image display surface 10, and other than the enlarged display portion. Description of this part is omitted.

The cylindrical lens length P L1 in the X-axis direction of the first cylindrical lens L S1 , the cylindrical lens length P L2 in the X-axis direction of the second cylindrical lens L S2 , and the cylindrical lens length P L3 in the X-axis direction of the third cylindrical lens L S3. The cylindrical lens lengths P L4 in the X-axis direction of the fourth cylindrical lens L S4 are all 5 sub-pixels when the 5-parallax method is adopted. Each cylindrical lens has a curvature in the X-axis direction and does not have a curvature in the Y-axis direction. The cylindrical lens length P L1 in the X-axis direction of the first cylindrical lens L S1 , the cylindrical lens length P L2 in the X-axis direction of the second cylindrical lens L S2 ,... Cylindrical lenses in the X-axis direction of all other cylindrical lenses It is a well-known technique to have the same length.

Distance between optical axes P CS1 from the optical axis of the first cylindrical lens L S1 to the optical axis of the second cylindrical lens L S2 , and light from the optical axis of the second cylindrical lens L S2 to the optical axis of the third cylindrical lens L S3 The inter-axis distance P CS2 and the inter-optical axis distance P CS3 from the optical axis of the third cylindrical lens L S3 to the optical axis of the fourth cylindrical lens L S4 are all equal. When the distances between the optical axes are equal, the distance between the optical axes is simply referred to as CS . The optical axis of the cylindrical lens is an axis in which light goes straight without being refracted. When each cylindrical lens is uniform, the cylindrical lens is positioned at an intermediate position in the X-axis direction length (X-axis direction length). There is an optical axis. That is, the axis along which the incident light goes straight without being refracted by the cylindrical lens is the optical axis. Distance between optical axes P CS1 from the optical axis of the first cylindrical lens L S1 to the optical axis of the second cylindrical lens L S2 , and light from the optical axis of the second cylindrical lens L S2 to the optical axis of the third cylindrical lens L S3 The distance P CS2 between the axes, the distance P CS3 between the optical axes from the optical axis of the third cylindrical lens L S3 to the optical axis of the fourth cylindrical lens L S4 , and the distance between the optical axes of the cylindrical lenses (not shown) are the same. Is a well-known technique.

Each of the first cylindrical lens L S1 , the second cylindrical lens L S2 , the third cylindrical lens L S3 , and the fourth cylindrical lens L S4 faces five subpixels in the X-axis direction. Such an arrangement is called a 5-parallax method. Also, the four-parallax method is used when each of the cylindrical lenses faces four sub-pixels in the X-axis direction, and the seven-parallax method is used when each of the cylindrical lenses faces seven sub-pixels in the X-axis direction. That is, the name of the parallax method is determined depending on how many sub-pixels arranged in the axial direction have an axial length having the curvature of the cylindrical lens. When each of the cylindrical lenses faces three sub-pixels in the X-axis direction, a combination of R, G, and B that can be viewed as one pixel by the viewer is not possible. Cannot be configured. In other cases, the stereoscopic image display device can be configured in principle even with other parallax methods, not limited to the 4-parallax method, the 5-parallax method, and the 7-parallax method.

FIG. 4 is a cross-sectional view at a stage where the basic stereoscopic image display optical element 11 is provided to the market as an optical component. FIG. 4 shows only a part of the stereoscopic image display optical element. Protective sheet, adhesive, base material, polymethyl methacrylate resin or acrylic resin (Poly methyl) made of polyethylene terephthalate (PET)
A lenticular lens composed of a plurality of cylindrical lenses made of (methacrylate: PMMA) is laminated and provided to the market as an optical component.

  In the step of using the stereoscopic image display optical element 11, the laminate of the adhesive, the substrate, and the lenticular lens is peeled off from the protective sheet. Then, a laminate of the substrate to which the adhesive is attached and the lenticular lens is used by being adhered to the image display surface 10 of the flat image display device. That is, the stereoscopic image display optical element 11 shown in FIGS. 3, 4, and 5 is a laminate of an adhesive, a substrate, and a lenticular lens.

The optical axis of the first cylindrical lens L S1 passes through the most convex part (hereinafter referred to as the most convex part) a 1 of the cylindrical lens, and the optical axis of the second cylindrical lens L S2 is the most convex part a 2 of the cylindrical lens. Pass through. That is, the most convex part exists on the optical axis. The portion where the first cylindrical lens L S1 and the second cylindrical lens L S2 are in contact is the most convex portion (the most concave portion) b 1 . Outermost protrusions a 1 (optical axis) and the length (the distance between the optical axes) P C1 between the outermost convex portion a 2 (optical axis) of the second cylindrical lens L S2 of the first cylindrical lens L S1 in principle the length P CS in FIG., the (n-1) cylindrical lenses L Sn-1 of the most projecting portion a n-1 and the length P Cn of the most projecting portion a n of the n cylindrical lens L Sn ( until not shown), all of the distance between the optical axes, in the principle diagram is constant length P CS. X-axis length of all the cylindrical lenses X-axis direction of the cylindrical lens length P L1 and other first cylindrical lens L S1, in the principle diagram the length P P, are all the same length.

FIG. 5 shows the positional relationship between the image display surface of the stereoscopic image display device and the stereoscopic image display optical element having a plurality of cylindrical lenses in the embodiment from a cross-sectional direction perpendicular to the image display surface (AA cross-sectional direction in FIG. 3). FIG. In FIG. 5, 5 sub-pixel length P P1 = 5 sub-pixel length P P2 = 5 sub-pixel length P P3 = cylindrical lens length P L1 = X-axis direction cylindrical lens length P L2 = X-axis direction Cylindrical lens length P L3 = P P.

FIG. 5 schematically shows the relationship between the sub-pixels of the stereoscopic image display device and the cylindrical lens. The distance LRL is the distance between the left eye and the right eye, and the average distance LRL has a size of 6.5 cm (centimeter). As illustrated in FIG. 4, the subpixels denoted by reference numerals 2 and 3 are subpixels visually recognized by the right eye, and the subpixels denoted by numerals 4 and 5 are subpixels visually recognized by the left eye. An image corresponding to the right eye is displayed on the subpixels denoted by reference numerals 2 and 3, and an image corresponding to the left eye is displayed on the subpixels denoted by reference numerals 4 and 5.

When the distance between the stereoscopic image display optical element 11 and the straight line connecting the left eye and the right eye is the distance L DZ1 , the subpixel labeled 2 is visually recognized by the right eye (see the dashed arrow labeled 2) ), The sub-pixels marked with 5 are viewed by the left eye (see dashed arrow marked with 5). In this way, the distance L DZ1 at which the viewer can recognize the stereoscopic image with the right eye and the left eye has a certain allowable width. In addition, when the separation distance between the stereoscopic image display optical element 11 and the straight line connecting the left eye and the right eye is the distance L DZ2 , the subpixel denoted by reference numeral 3 is visually recognized by the right eye (the broken line arrow denoted by reference numeral 3). ), The sub-pixels marked 4 are visually recognized by the left eye (see dashed arrows marked 4). The distance L DZ2 at which the viewer can recognize the stereoscopic image has a certain allowable width.

In this way, when the separation distance between the stereoscopic image display optical element 11 and the straight line connecting the left eye and the right eye is in the vicinity of the distance L DZ1 , the nearest three sub-colors R, G, and B denoted by reference numeral 2 are attached. The pixels from the entire image display surface can be visually recognized by the right eye so that the pixels constitute one pixel. Further, the pixels from the entire image display surface can be visually recognized by the left eye so that the latest three sub-pixels of R, G, and B denoted by reference numeral 5 constitute one pixel. And the image visually recognized by the right eye and the image visually recognized by the left eye are arranged in advance as those recognized by a human as a stereoscopic image.

Similarly, when the separation distance between the stereoscopic image display optical element 11 and the straight line connecting the left eye and the right eye is in the vicinity of the distance L DZ2 , the nearest three subpixels of R, G, and B denoted by reference numeral 3 are used. Constitutes one pixel, and pixels from the entire image display surface can be viewed with the right eye. Further, the pixels from the entire image display surface can be visually recognized by the left eye so that the latest three subpixels of R, G, and B denoted by reference numeral 4 constitute one pixel. And the image visually recognized by the right eye and the image visually recognized by the left eye are arranged in advance as those recognized by a human as a stereoscopic image.

If the stereoscopic image recognized in the vicinity of the distance L DZ1 and the stereoscopic image recognized in the vicinity of the distance L DZ2 have the same content, the viewer can recognize the stereoscopic image display optical element 11 and the left eye and Even if the distance between the straight line connecting the right eye and the right eye changes greatly, a stereoscopic image having the same content can be recognized. Further, if the content of the three-dimensional image are different recognized in the vicinity of the separation distance distance between the three-dimensional image to be recognized in the vicinity of the distance of the distance L DZ1 L DZ2, a viewer, a stereoscopic image display optical element 11 When the distance between the left eye and the straight line connecting the right eye changes greatly, stereoscopic images with different contents can be recognized.

  In the technique shown in the principle of the stereoscopic image display apparatus described with reference to FIGS. 1 to 5 described above, so-called 3D moire (three-day moire) is generated and the quality of the stereoscopic image is impaired. The principle of 3D moire generation is also described in Patent Document 4 (Japanese Patent Laid-Open No. 2005-208567).

(Principle of 3D moire reduction)
The inventor (referred to as the inventor) described in the application of the present application has conducted extensive research on the removal of 3D moire, and has confirmed that 3D moire can be reduced by various embodiments described below. The 3D moire mitigation technique reached by the inventors is summarized as follows.

First, the place where the inventor recognizes the cause of occurrence of 3D moire is summarized.
(1) The cause of the occurrence of 3D moire is that light that has passed through each cylindrical lens functioning as a diffraction grating interferes regularly.
(2) As each cylindrical lens of the stereoscopic image display optical element 11 is more uniform, the light interference is visually recognized as a thick black line. This is 3D moire. Here, that the cylindrical lenses are uniform means that the lengths of the respective cylindrical lenses in the X-axis direction are equal, and as a result, the distance between the optical axes of adjacent cylindrical lenses is also equal.

Next, the main points of the invention made by the inventor regarding the 3D moire reduction method will be summarized.
(1) If the uniformity of each cylindrical lens of the stereoscopic image display optical element 11 is broken so that the light passing through each cylindrical lens functioning as a diffraction grating does not interfere regularly, the light passing through each cylindrical lens is regular. Will not be recognized as a thick black line. That is, 3D moire can be reduced.
(2) The problem is how to reduce the 3D moire without affecting the image quality of the stereoscopic image (stereoscopic image quality) by breaking the uniformity of each cylindrical lens.
(3) The inventor has found that the image quality of a stereoscopic image is determined by two indexes of image quality and stereoscopic characteristics. Here, the image quality is an index as to whether or not R, G, and B appear to be appropriately mixed as one pixel regardless of whether or not it looks three-dimensional. The stereoscopic characteristic is an index as to whether or not light from a sub-pixel is incident on each of the left and right eyes of the viewer at an angle causing a stereoscopic effect.
B)
Regarding the image quality, the refraction direction of light changes critically at the boundary between the nth cylindrical lens and the (n + 1) th cylindrical lens (the most concave portion b n ), so the end of the light emitting portion (light emitting portion) of the subpixel emits light. The image quality deteriorates when this part faces the position that is largely inside from the side.
B)
Regarding the stereoscopic characteristics, a distance difference (position) between a position of a partition of a plurality of sub-pixels on the image display surface 10 (for example, a position of a partition for every five sub-pixels in the 5-parallax method) and the most concave portion of the cylindrical lens. When the (phase difference) is shifted within one subpixel, the stereoscopic characteristics of the stereoscopic image recognized by the viewer in accordance with the shift amount deteriorate. If this phase difference is shifted by one or more subpixels, the combination of subpixels constituting the pixels of the stereoscopic image recognized by the viewer is shifted, so that the stereoscopic effect is impaired and the stereoscopic characteristics are remarkably deteriorated.
(4) Therefore, in view of the index of (3), if the uniformity of each cylindrical lens is appropriately destroyed, 3D moire can be reduced with almost no deterioration in the quality of the stereoscopic image.
In the following, an embodiment that specifically destroys the uniformity of each cylindrical lens will be described.

(First embodiment)
The stereoscopic image display optical element of the first embodiment will be described with reference to FIGS. The stereoscopic image display optical element according to the first embodiment is an embodiment in which the uniformity of the distance between optical axes of adjacent cylindrical lenses is lost.

  FIG. 6 is a view of the positional relationship between the image display surface 10 of the stereoscopic image display device according to the first embodiment and the stereoscopic image display optical element 11A having a plurality of cylindrical lenses, viewed from the image display surface direction.

  In FIG. 6, in order to make the drawing easy to see, the description of the light shielding portion between the sub-pixels is omitted, and the most concave portion that is the boundary of each cylindrical lens is emphasized and written thickly. The stereoscopic image display optical element 11A displays only the enlarged display portion, but the stereoscopic image display optical element 11A is disposed so as to face the entire surface of the image display surface 10, and other than the enlarged display portion. Description of this part is omitted.

The image display surface 10 shown in FIG. 6 has the same shape as the image display surface 10 shown in FIGS. In the first embodiment, the five-parallax method will be described, but the technical idea does not lose generality in other parallax methods. The five subpixel lengths are described as P P1 , P P2 and P P3 . All five consecutive subpixels (5 subpixel length) arranged on the image display surface are equal. That is, since it is P P1 = P P2 = P P3 ···, will be described below 5 sub pixel length by replacing the P P.

The stereoscopic image display optical element 11A shown in FIG. 6 has a different shape from the known stereoscopic image display optical element 11 shown in FIGS. That is, the stereoscopic image display optical element 11A shown in FIG. 6, the outermost protrusion a 1 and the optical axis distance P CS1 is a distance between the outermost convex portion a 2 of the second cylindrical lens L S2 of the first cylindrical lens L S1 , the distance between the optical axes P CS3 is an interval between the outermost convex portion a 3 of the outermost convex portion a 2 and the third cylindrical lens L S3 of the second cylindrical lens L S2, top protrusions a 3 of the third cylindrical lens L S3 as each optical axis distance P CS4 is an interval between the outermost convex portion a 4 is the fourth cylindrical lens L S4, not uniform. Written by formula, where n is an arbitrary integer, most convex part a n and the (n + 1) cylindrical lenses L Sn + 1 of the distance between the outermost convex portion a n + 1 is inter-optical axis distance P CSn of the n cylindrical lens L Sn However, the distance P CSn between the optical axes is not uniform.

  First, the first embodiment will be described using a general formula.

As described above, with reference to the coordinate axis on the image display surface 10, the distance L Cn from the X-axis origin (X = 0) to the most concave portion b n of the n-th cylindrical lens constituting the stereoscopic image display optical element 11A is: It is represented by the general formula of Formula 1. In Equation 1, it is assumed that the length in the axial direction (X-axis direction) having the curvature of the n-th cylindrical lens L Sn is P Ln , and the most concave portion b 0 of the first cylindrical lens L S1 is located at the origin. To do.

The distance L Cn represented by Equation 1 is the distance of the boundary between the nth cylindrical lens and the (n + 1) th cylindrical lens from the origin.

On the other hand, the distance from the X-axis origin (X = 0) of the five subpixels in units of five subpixels on the image display surface 10 is the distance L P1 (not shown) to the fifth subpixel. Is P P , the distance L P2 (not shown) to the 10th sub-pixel is P P + P P = 2 × P P , and the distance L P3 (not shown) to the 15th sub-pixel is P P1 + P P2 + P P3 = 3 × P P

A distance L Pn from the origin of the X axis to the (5 × n) th sub-pixel of the image display surface 10 is expressed by Equation 2.

The distance L Pn expressed by Equation 2 is the distance from the origin to the boundary between the (5 × n) th subpixel and the {(5 × n) +1} th subpixel.

Each of the (5 × n) -th subpixel and the {(5 × n) +1} -th subpixel has a light shielding portion having a length P SX . At the boundary between the (5 × n) th subpixel and the {(5 × n) +1} th subpixel, the length of the light shielding portion is 2 × P SX .

At the boundary between the n-th cylindrical lens and the (n + 1) -th cylindrical lens (the most concave portion b n ), the light refraction direction changes critically, so that the sub-pixel emits light at the most concave portion b n. When faced, image quality deteriorates. Therefore, in order to make the most concave portion b n within the range facing the light shielding portion, the phase difference must be within the range represented by Equation 3. Equation 3 is a conditional expression for the cylindrical lens's most concave portion b n to remain within the range of the light shielding portion at the boundary between the (5 × n) th subpixel and the {(5 × n) +1} th subpixel.

Further, the distance difference (phase difference) between the distance L Cn on the stereoscopic image display optical element 11A from the origin 0 in FIG. 6 and the distance L Pn on the image display surface 10 from the origin 0 in FIG. If the pixel is shifted by more than one pixel, the combination of the sub-pixels constituting the pixel of the stereoscopic image viewed in each arrow direction (see FIG. 5) is shifted, so that the stereoscopic characteristic is deteriorated and the stereoscopic image cannot be seen. Also, the image quality deteriorates. Therefore, the phase difference must be within a range within one subpixel, but the range is determined by experiment. Formula 4 is a formula that defines an allowable range of deterioration of the stereoscopic characteristics and image quality. In Equation 4, k is a constant equal to or less than 1, and is a value determined by experiment. The value of k determines an allowable range of deterioration in stereoscopic characteristics and image quality.

ABSs in Equations 3 and 4 are absolute values. Was to obtain an absolute value for the phase difference between the boundary between the n-th cylindrical lens L Sn outermost recess b n and reference numeral 5 and the code 1 subpixels, the phase lead (L Pn -L Cn> 0) This is because both the phase delay and the phase delay (L Pn −LC n <0) are considered.

Since the sub-pixel length P SPX in the X-axis direction of each sub-pixel includes the light shielding portion length P SX in the X-axis direction, P SPX > P SX . Here, P SPX = P P / NSH in Formula 4. NSH is an integer corresponding to the parallax method, 4 for the 4-parallax method, 5 for the 5-parallax method, and 7 for the 7-parallax method. That is, there is a relationship of the length P SPX = P P / NSH of the subpixels facing in the direction having the curvature of the cylindrical lens. The meaning of k and the value of k in Equation 4 will be described later, but k is a number less than 1 (k <1).

  It has already been described that the degradation of the stereoscopic image quality is caused by the degradation of the image quality and the degradation of the stereoscopic characteristics. As described above, in order for the degradation of the stereoscopic image quality to be within the allowable range, the degradation of the image quality must be within the allowable range and the degradation of the stereoscopic characteristics must be within the allowable range. The main points between the deterioration of the stereoscopic image quality and the relationship between Equations 3 and 4 are summarized below.

  The k in Equation 4 will be described. As described above, when k ≧ 1, the deterioration of the stereoscopic characteristics is remarkably deteriorated. Even if k <1, degradation of stereoscopic characteristics and degradation of image quality occur, so that the value of k is allowed to be subjected to a visual experiment by a plurality of subjects, and degradation of the stereoscopic characteristics and degradation of image quality are allowed. It is defined as possible. According to the inventors' experiment, the value of k was less than 0.25.

How to determine the sub-pixel length P SPX and the light shielding portion length P SX of the sub-pixel is a matter that can be widely determined as a design matter. The inventors have actually measured the dimensions of the subpixels arranged on the image display surface of a flat image display device on the market, and the light shielding portion length PSX and the length of the light emitting portion in the subpixel length PSPX are as follows. The ratio with the light emitting part length (P SPX -P SX ) is widely distributed. Similarly, the ratio between the light emitting director (P SPY -P SY) is the length of the portion that emits light shielding director P SY and the light in the sub-pixel length P SPY is distributed widely. Here, C X = P SX / P SPX and C Y = P SY / P SPY . In the first embodiment, P SX , P SPX , and C X have a meaning, but in the two-parallax method described later, P SY , P SPY , and CY have a meaning.

The inventor has actually measured the dimensions of the sub-pixels of the image display devices of each company on the market. In the image display apparatus of company A, P SX : (P SPX −P SX ) = 0.033: 0.967 and P SY : (P SPY −P SY ) = 0.13: 0.87. At this time, C X = 0.033 and C Y = 0.13. In the image display apparatus of company B, P SX : (P SPX −P SX ) = 0.35: 0.65 and P SY : (P SPY −P SY ) = 0.15: 0.85. At this time, C X = 0.35 and C Y = 0.15. In the image display device of company C, P SY : (P SPY −P SY ) = 0.036: 0.964 and P SX : (P SPX −P SX ) = 0.09: 0.91. At this time, C X = 0.036 and C Y = 0.09. Thus, P SX of the image display device of each company in the market: (P SPX -P SX) ratio of, P SY: (P SPY -P SY) ratio of, C X, various of the C Y There is something.

(Case 1) When k ≦ C X = P SX / P SPX , k in Equation 4 is set to a range where degradation of stereoscopic characteristics and degradation of image quality are acceptable, that is, a range where stereoscopic image quality is acceptable. . In the case of k = C X , Equation 3 and Equation 4 are the same mathematical expressions. Therefore, in the case of k ≦ C X , if Equation 4 is established, Equation 3 is necessarily satisfied. When Equation 3 is satisfied, the most concave portions of all the cylindrical lenses are located in the range of the light shielding portion length P SX , and the image quality hardly deteriorates. That is, in the case of k ≦ C X , if the expression 4 is satisfied, the stereoscopic image quality is in an acceptable range. However, if the expression 3 is satisfied, a further excellent stereoscopic image quality can be obtained. Become.

(Case 2) When k> C X = P SX / P SPX The k of Equation 4 determines the range in which the degradation of the stereoscopic characteristics and the degradation of the image quality can be tolerated. In the case of k> C X , even if Equation 4 is satisfied, Equation 3 is not satisfied. Since the condition of Equation 4 is satisfied, it is within the range of acceptable degradation of stereoscopic characteristics and within the range of acceptable degradation of image quality. That is, in the case of k> C X , even if the most concave portion of the cylindrical lens is positioned beyond the range of the light shielding portion length P SX , if the expression 4 is satisfied, the stereoscopic image quality is within an allowable range.

As described above, in the case of (Case 1), if the expression 3 is satisfied, the image quality hardly deteriorates, so that it is more desirable than the case of (Case 2). Which case corresponds to (Case 1) or (Case 2) is determined by the structure of the subpixel. In either case (Case 1) or (Case 2), Equation 4 indicates an index of how much degradation of the stereoscopic image quality can be tolerated no matter how large ABS [L Pn −L Cn ] is set for each cylindrical lens. It becomes the mathematical formula for. On the other hand, the inventor's experiment confirmed that moire has a sufficient moire reduction effect without increasing ABS [L Pn -L Cn ] up to k × P SPX which is the maximum limit shown in Equation 4. Yes.

Note that the stereoscopic image display device of the principle diagram (see FIGS. 3, 4 and 5) has a constant length P Ln = P P and n in the direction having the curvature of all the cylindrical lenses, and the distance This is a special case of Formula 4 in which the distance difference (phase difference) between L Cn and the distance L Pn is zero. Equation 5 is established when k = 0 in Equation 4. Equation 5 is a mathematical formula that holds in the case of the principle diagram.

Under the condition of Equation 5, since ABS [L Pn -L Cn ] = 0 for all the cylindrical lenses, the image quality of the stereoscopic image does not deteriorate at all, but the moire reduction effect does not occur at all.

In the first embodiment shown in FIG. 6, when the length types of the cylindrical lenses are preliminarily determined so as to satisfy the conditional expressions shown in Equations 3 and 4, as few cylindrical lenses as possible (types of lengths of the cylindrical lenses) The stereoscopic image display optical element 11A having two types of cylindrical lenses of length P LA and length P LB will be described below. P LA in parentheses below the P L1, P L4 of FIG. 6, will be described below with reference to P LB in parentheses below the P L2, P L3.

By arranging two cylindrical lenses of length P LA and two cylindrical lenses of length P LB alternately in the X-axis direction, the light of adjacent cylindrical lenses is lost while breaking the uniformity of the length of each cylindrical lens. An explanation will be given of the case where the inter-axis distance is not uniform. In this case, satisfying Equation 4 can make the stereoscopic image quality acceptable. In addition, when the subpixel structure satisfies the condition of k ≦ CX , if the expression 3 is satisfied, a better stereoscopic image quality can be obtained.

A description will be given with reference to FIG. 6 assuming that P LA = P P + δ and P LB = P P -δ. Here, P P is a length that satisfies Equation 5 corresponding to the case where there is no variation in the length of the cylindrical lens in the axial direction (X-axis direction) having the curvature, and is referred to as a reference length P P. The origin (X = 0) is the position of the extreme end of the image display surface 10 and is the end of R (red) with reference numeral 1. Position of arrangement of the first cylindrical lens L S1 is to match the case of 4 n = 1, the the description below for the case where the outermost recess b 0 of the cylindrical lens L S1 to the origin is located.

The distance L C1 from the origin to the most concave part b 1 of the cylindrical lens is P P + δ. Distance L C2 from the origin to the top recess b 2 is 2 × P P. The distance L C3 from the origin to the most concave portion b 3 is 3 × P P −δ. Distance L C4 from the origin to the top recess b 4 is a 4 × P P. Although not shown, the distance L C5 is 5 × P P + δ, the distance L C6 is 6 × P P , the distance L C7 is 7 × P P −δ, and so on, after the most concave portion b 5. . Here, the positions of the ends of the subpixels denoted by reference numeral 5 are P P , 2 × P P , 3 × P P , 4 × P P , 5 × P P , 6 × P P , 7 × P P,.・ ・. Therefore, the most concave portion of the cylindrical lens with respect to the end of the subpixel denoted by reference numeral 5 sequentially repeats + δ phase shift, no phase shift, and −δ phase shift (see Formulas 3 and 4).

Also, the position of the uppermost recess (see Figure 3) when all of the cylindrical lens is a reference length P P, length P LA of the cylindrical lens, P LB, P LB, P LA, P LA, P LB , P LB , P LA , P LA ..., And the phase difference from the position of the most concave portion (see FIG. 6) are repeated in order of + δ phase shift, no phase shift, and −δ phase shift. .

The distance from the origin (X = 0) to the most convex part a 1 that is the optical axis is 0.5 × P P + 0.5 × δ. Distance from the origin top to protrusion a 2 is the optical axis is 1.5 × P P. Distance from the origin to the highest protrusion a 3 is the optical axis is 2.5 × P P -0.5 × δ. Distance from the origin top to protrusion a 4 is the optical axis is 3.5 × P P. Although not shown, as below Saitotsu unit a 5 or later, 4.5 × P P + 0.5 × δ, 5.5 × P P, 6.5 × P P -0.5 × δ, 7. 5 × P P ...

The optical axis distance P CS1 is P P. The inter-optical axis distance P CS2 is P P + δ. The optical axis distance P CS3 is P P. Although not shown, the optical axis distance PCS4 is P P -δ. Inter-optical axis distance P CS5 later, P P, P P + δ , P P, P P -δ, repeated P P · · · and order. In this way, there are three types of optical axis distances.

In the case of (Case 1), there is no reduction in image quality even if there is a phase shift up to the length of the light shielding part length P SX in the X-axis direction as shown in Equation 3. When a maximum difference is made between the length of P LA and the length of P LB within a range in which the image quality is not reduced, δ = P SX and P LA = P P + P SX , P LB = By setting to P P -P SX , a very good image quality can be obtained. In the case of (Case 2), as shown in Equation 4, if δ is appropriately selected within the range of δ <k × P SPX , the degradation of the stereoscopic image quality can be allowed. There is a relationship of sub-pixel length P SPX = P P / NSH, and according to experiments conducted by the inventors, the value of k was less than 0.25.

  Two methods for manufacturing the above-described stereoscopic image display optical element having cylindrical lenses having two different cylindrical lengths and three different distances between optical axes will be described below.

In the first method, the value of the radius of curvature R shown in FIG. 4 is made different between a cylindrical lens having a length P LA and a cylindrical lens having a length P LB. Cylindrical lens length P LA (e.g., a first cylindrical lens L S1) to vary the value of the curvature radius R 2 of curvature R 1 and the length P LB of the cylindrical lens (e.g., the second cylindrical lens L S2) of . In this way, a stereoscopic image display optical element can be configured by three types of cylindrical lenses having different optical axis distances. In this case, the position of the most convex part may be the same position in the Z-axis direction, and the position of the most concave part may be the same position in the Z-axis direction. Further, both the position of the most convex part and the position of the most concave part in the Z-axis direction may be slightly adjusted. When the values of the radius of curvature of the cylindrical lens are two, that is, the radius of curvature R 1 and the radius of curvature R 2 , the stereoscopic image display optical element manufacturing mold has two types having the same curvature as the curvature of each cylindrical lens. The cutting tool is switched, and two types of grooves, a groove having a curvature corresponding to a cylindrical lens having a length P LA and a groove having a curvature corresponding to a cylindrical lens having a length P LB , are cut and manufactured. Can do.

In the second method, the value of the radius of curvature R shown in FIG. 4 is set based on a cylindrical lens having a length P LA (for example, a first cylindrical lens L S1 ) and a cylindrical lens having a length P LB (for example, a second cylindrical lens L S2 ) is the same. In this case, for example, the distance in the Z-axis direction from the image display surface 10 of the most convex portion a 1 (optical axis) of the first cylindrical lens and the image of the most convex portion a 2 (optical axis) of the second cylindrical lens. The distance from the display surface 10 in the Z-axis direction is slightly different. Further, the position of the most concave portion in the Z-axis direction may be slightly adjusted, and both the position of the most convex portion and the position of the most concave portion in the Z-axis direction may be slightly adjusted. In this way, the three-dimensional image display optical element 11A can be configured by three types of cylindrical lenses having different optical axis distances. When the value of the radius of curvature R of the cylindrical lens is one, the stereoscopic image display optical element manufacturing die switches the position in the Z-axis direction of one type of cutting tool to two types, and the length P LA can be manufacturing by cutting the two grooves of the cylindrical grooves corresponding to lenses and the length P LB cylindrical lenses corresponding groove in the.

  The above-described first method or the above-described second method can be extended to a method for manufacturing a stereoscopic image display optical element having any kind of cylindrical lens. When many types of concave grooves are cut into a mold using the first method, the same number of cutting tools as the number of types of concave grooves are required, and the cost of mold manufacture increases. When many types of concave grooves are cut into a mold using the second method, the cost of mold manufacture can be reduced because only the position in the Z-axis direction is switched so as to be many types.

  FIG. 7 is a cross-sectional direction perpendicular to the image display surface (cross section AA in FIG. 6) showing the positional relationship between the image display surface of the stereoscopic image display device according to the first embodiment and a stereoscopic image display optical element having a plurality of cylindrical lenses. FIG.

In FIG. 7, the relationship between the pixels of the stereoscopic image display device and the cylindrical lens is schematically shown by a plan view. When FIG. 7 and FIG. 5 are compared, the dimensions relating to the image display surface 10 are the same. That is, the sub-pixel length P SPX in the X-axis direction of one sub-pixel and the light-shielding part length P SX in the X-axis direction of one sub-pixel are the same in FIGS.

On the other hand, as described above, for the dimensions related to the stereoscopic image display optical element 11A shown in FIG. 7, for example, the length of the cylindrical lens in the X-axis direction is different every two. Sum of the lengths of two different cylindrical lenses is twice the length P P. This is different from the stereoscopic image display optical element 11 shown in FIG. 5 in which the cylindrical lens has the same length in the X-axis direction. Further, as shown in Equations 3 and 4, in general, the type of length of the cylindrical lens is not limited to two.

In the vicinity of the distance L DZ1 between the stereoscopic image display optical element 11A and the straight line connecting the left eye and the right eye, the nearest three subpixels of R, G, and B denoted by reference numeral 2 represent one pixel. It can be visually recognized by the right eye as configured, and can be visually recognized by the left eye so that the latest three sub-pixels of R, G, and B denoted by reference numeral 5 constitute one pixel. Further, when the separation distance is in the vicinity of the distance L DZ2 , the nearest three subpixels of R, G, and B with reference numeral 3 can be visually recognized by the right eye so as to form one pixel. The latest three subpixels of R, G, and B with 4 can be visually recognized by the left eye so as to form one pixel. The point that the image visually recognized by the right eye and the image visually recognized by the left eye are recognized as a stereoscopic image by the human is the same as the principle described with reference to FIG.

  When the stereoscopic image display optical element 11A shown in FIGS. 6 and 7 is used, there are three types of distances between the optical axes of adjacent cylindrical lenses of the stereoscopic image display optical element 11A, and there are two types of lengths of the cylindrical lenses. Since different low-frequency components are generated in the light moire from each cylindrical lens, the apparent degree of occurrence of 3D moire is reduced as a result. In general, the length of the axial length having the curvature of the cylindrical lens is limited to two, and as a result, the distance between the optical axes is not limited to three. As shown in Equations 3 and 4, any kind of cylindrical lens can be used. The degree of occurrence of 3D moiré is reduced as the number of axial lengths having the curvature of the cylindrical lens increases.

(Modification of the first embodiment)
In the first embodiment, description has been made assuming that the length of the cylindrical lens is varied, and as a result, the distance between the optical axes is varied. Therefore, it can be considered that 3D moire is reduced by changing the distance between the optical axes of the cylindrical lenses. The modification of the first embodiment described below is an embodiment from the viewpoint of making the distance between the optical axes of the cylindrical lenses various.

Description will be made with reference to FIG. 6 again. When paying attention to the distance between the optical axes of the cylindrical lenses, Equation 1 can be rewritten into another expression method. As shown in Equation 6, PCS0 is ½ of the cylindrical lens length P L1 in the X-axis direction of the cylindrical lens L S1 . P CSn corresponding to an integer n of 1 or more is represented by the sum of ½ of the length in the X-axis direction of two adjacent cylindrical lenses.

The distance L Cn represented by Expression 7 is the distance of the boundary between the n-th cylindrical lens and the (n + 1) -th cylindrical lens from the origin, and is another expression of Expression 1.

  Another expression of Equation 3 using Equation 7 is Equation 8. Another expression of Equation 4 using Equation 7 is Equation 9.

In Expressions 6 and 7, the optical axis distance P CSn is clearly shown, but in Expressions 8 and 9, the optical axis distance P CSn is not used, but the cylindrical lens length P Ln in the X-axis direction of the cylindrical lens. It is expressed. Thus, the making fluctuated a length in the X-axis direction of the cylindrical lens length P Ln of sand Lind helical lens, essential difference is not between the letting fluctuated the distance between the optical axes of adjacent cylindrical lenses. In terms of the mathematical expression, the distance between the optical axes P CSn is expressed by the cylindrical lens length P Ln in the X-axis direction of the sub-pixel. There is no substantial difference from 4. As shown in Equations 6 and 7, in the modification of the first embodiment, the distance P CSn between the optical axes of the cylindrical lenses is an arbitrary type. The degree of occurrence of 3D moire is reduced as the types of inter-optical axis distances PCSn increase. There is a relationship of sub-pixel length P SPX = P P / NSH, and according to experiments conducted by the inventors, the value of k was less than 0.25.

  (Second Embodiment)

The second embodiment is the same as the first embodiment in that it features a stereoscopic image display optical element having cylindrical lenses having a plurality of predetermined lengths. However, the following points are different between the second embodiment and the first embodiment. Distance between optical axes P CS1 from the optical axis of the first cylindrical lens L S1 to the optical axis of the second cylindrical lens L S2 , and light from the optical axis of the second cylindrical lens L S2 to the optical axis of the third cylindrical lens L S3 The inter-axis distance P CS2 , the optical axis distance P CS4 from the optical axis of the third cylindrical lens L S3 to the optical axis of the fourth cylindrical lens L S4 , in the general formula, the (n−1) th cylindrical lens L Sn−1 The optical axis distance P CSn between the optical axis and the optical axis of the n-th cylindrical lens L Sn is made equal regardless of the value of n. All values of the optical axis distance P CSn corresponding to an arbitrary n are P CS = P P.

  FIG. 8 is a view of the positional relationship between the image display surface 10 of the stereoscopic image display apparatus according to the second embodiment and the stereoscopic image display optical element 11B having a plurality of cylindrical lenses, viewed from the image display surface direction.

  In FIG. 8, in order to make the drawing easier to see, the description of the light shielding portion between the sub-pixels is omitted, and the boundary of each cylindrical lens is emphasized and written thickly. In the stereoscopic image display optical element 11B, only the enlarged display portion is displayed. The stereoscopic image display optical element 11B is disposed so as to face the entire surface of the image display surface 10, and description of other parts other than the enlarged display portion is omitted.

  The image display surface 10 shown in FIG. 8 has the same shape as the image display surface 10 shown in FIGS.

Figure 8 is a top protrusion a 1 is the most convex portion of the cylindrical lenses, Saitotsu section a 2, Saitotsu unit a 3, the position of each of the Saitotsu unit a 4, the sub-pixels denoted by reference numeral 3 The case where it coincides with the center position is shown.

The distance L C1 from the X-axis origin (X = 0) to the most concave portion b 1 of the first cylindrical lens constituting the stereoscopic image display optical element 11B (see FIGS. 8 and 9) is P P + 0.5 × It is represented by (P L1 -P P ). A distance L C2 from the X-axis origin (X = 0) to the most concave portion b 2 of the second cylindrical lens is represented by 2 × P P + 0.5 × (P L2 −P P ). A distance L C3 from the origin of the X axis (X = 0) to the most concave portion b 3 of the third cylindrical lens is represented by 3 × P P + 0.5 × (P L3 −P P ). A distance L C4 from the origin of the X axis (X = 0) to the most concave portion b 4 of the fourth cylindrical lens is represented by 4 × P P + 0.5 × (P L4 −P P ).

The following is written using a general formula. The distance from the uppermost recess b n-1 to the lowest recess b n length of the n cylindrical lenses of the n cylindrical lens is defined as P Ln. A distance L Cn from the X-axis origin (X = 0) to the most concave portion b n of the n-th cylindrical lens constituting the stereoscopic image display optical element 11B is expressed by Equation 6. The distance L Cn represented by Equation 6 is the distance from the origin to the boundary between the nth cylindrical lens and the (n + 1) th cylindrical lens.

On the other hand, the distance from the X-axis origin (X = 0) of the sub-pixel in units of five sub-pixels on the image display surface 10 is the distance L P1 (not shown) to the fifth sub-pixel. P P , the distance L P2 (not shown) to the tenth subpixel is P P + P P = 2 × P P , and the distance L P3 (not shown) to the fifteenth subpixel is P P1 + P P2 + P P3 = 3 × P P , and a distance L P4 (not shown) to the 20th subpixel is expressed as P P1 + P P2 + P P3 + P P4 = 4 × P P.

The distance L Pn from the origin of the X axis to the (5 × n) -th subpixel of the image display surface 10 is expressed by the above-described formula 2. The distance L Pn expressed by Equation 2 is a boundary between the (5 × n) th subpixel and the {(5 × n) +1} th subpixel.

At the boundary between the nth cylindrical lens and the (n + 1) th cylindrical lens (the most concave part b n ), the direction of light refraction changes critically, so this part faces the light emitting part (light emitting part) of the subpixel. Then, the image quality deteriorates.

Equation 11 is a conditional expression for the most concave portion b n to remain in the range of the light shielding portion at the boundary between the (5 × n) th subpixel and the {(5 × n) +1} th subpixel.

In addition, when the distance difference (phase difference) between the distance L Cn and the distance L Pn is shifted by one or more subpixels, a stereoscopic image that is visually recognized in each arrow direction (see FIG. 5) denoted by reference numerals 2 to 5. As a result, the combination of the sub-pixels constituting the pixels constituting the image is shifted, the stereoscopic characteristics are deteriorated, and the stereoscopic image cannot be seen. Therefore, the phase difference must be within a range within one subpixel, but the range is determined by a visual experiment by a plurality of subjects. According to the experiment conducted by the inventors, k in Equation 12 was less than 0.25, as in the first embodiment.

The meaning of Equation 11 is that, as described above with respect to Equation 3, even if the position of the most concave portion of each cylindrical lens is deviated within the range of P SX from the original position, the image quality is not affected. . Further, the expression (12) means that, as described above with respect to the expression (4 ), an index indicating whether degradation of the stereoscopic image quality can be tolerated no matter how large ABS [L Pn -L Cn ] is set for each cylindrical lens. It is a numerical formula for showing. Further, P SPX = P P / NSH. As described above, the value of NSH is 5 in the 5-parallax method. Other parallax methods correspond to integers corresponding to the parallax method.

Equations 11 and 12 indicate that the optical axis distance P CSn between the optical axis of the (n−1) th cylindrical lens L Sn-1 and the optical axis of the nth cylindrical lens L Sn is independent of the value of n. Although it is a general formula in the case of being equal, the more limited example in 2nd Embodiment is demonstrated. In FIG. 8, three-dimensional images in the case of using two types of cylindrical lenses with different lengths in the curvature direction, such as P L1 = P LA , P L2 = P LB , P L3 = P LA , P L4 = P LB. The display optical element 11B will be described below.

The description will be given with reference to FIG. 8 again. The cylindrical lens length P L2n-1 in the X-axis direction of the (2n-1) cylindrical lens L S2n-1 of the stereoscopic image display optical element 11B shown in FIG. 8 is the length P LA , and the second n cylindrical lens L S2n It is assumed that the cylindrical lens length P L2n in the X-axis direction is a length P LB. n is a positive integer.

A description will be made with reference to FIG. 8 assuming that P L2n−1 = P LA = P P + δ and P L2n = P LB = P P −δ.

Distance from the origin top to protrusion a 1 is the optical axis is 0.5 × P P. Distance from the origin top to protrusion a 2 is the optical axis is 1.5 × P P. Distance from the origin to the highest protrusion a 3 is the optical axis is 2.5 × P P. Distance from the origin top to protrusion a 4 is the optical axis is 3.5 × P P. Although not shown, as below Saitotsu unit a 5 or later, 4.5 × P P, 5.5 × P P, 6.5 × P P, 7.5 × P P, 8.5 × P P and in turn increased by P P. The distance P CS between the optical axes of all the cylindrical lenses is P P.

The distance from the origin to the most concave portion b 1 is P P + 0.5 × δ. The distance from the origin to the most concave portion b 2 is 2 × P P −0.5 × δ. The distance from the origin to the most concave portion b 3 is 3 × P P + 0.5 × δ. The distance from the origin to the most concave portion b 4 is 4 × P P + 0.5 × δ. Although not shown, after the most concave portion b 5 , 5 × P P + 0.5 × δ, 6 × P P −0.5 × δ, 7 × P P + 0.5 × δ, 8 × P P as follows: Repeat as -0.5 × δ, 9 × P P + 0.5 × δ. The phase difference between the boundary between the subpixel denoted by reference numeral 5 and the subpixel denoted by reference numeral 1 on the image display surface 10 and the most concave portion of the cylindrical lens is a phase shift of + 0.5 × δ, −0.5 × δ The phase shift is repeated.

Depending on the structure of the sub-pixel, in the case of (Case 1), image quality is reduced even if there is a phase shift up to the length of the light-shielding part length P SX in the direction of curvature (X-axis direction) as shown in Equation 11. Therefore, when making the maximum difference between the length of P LA and the length of P LB , δ = P SX and P LA = P P + P SX , P LB = P P −P SX Can be set. If δ is appropriately selected within the range of δ ≦ P SX , the image quality is not reduced.

As shown in Equation 12, even if there is a phase shift up to the length of the light shielding portion P SX in the k × X axis direction, the stereoscopic characteristics and the image quality are not reduced. The value of k was less than 0.25 according to the experiment as described above. When making the maximum difference between the length of P LA and the length of P LB , set δ = k × P SPX and set P LA = P P + P SX and P LB = P P -P SX it can. It should be noted that, as described above, which of Equations (11) and (12) imposes a limit on the length of the cylindrical lens in the X-axis direction is, as described above, the subpixel of the image display surface 10 combined with the stereoscopic image display optical element 11B. Depends on the structure.

  Two methods for manufacturing a stereoscopic image display optical element having cylindrical lenses having two different lengths as described above will be described below.

In the third method, the value of the radius of curvature R shown in FIG. 4 is made different between the (2n-1) cylindrical lens L S2n-1 and the second n cylindrical lens L S2n . The (2n-1) cylindrical lenses L S2n-1 (e.g., a first cylindrical lens L S1) of curvature radius R 1 and the 2n cylindrical lens L S2n (e.g., the second cylindrical lens L S2) of the radius of curvature R 2 of Set the value to satisfy the following conditions. The same position position as shown in FIG. 4 and the X-axis direction of the outermost convex portion a 2 (optical axis) of the top protrusion a 1 (optical axis) and a second cylindrical lens L S2 of the first cylindrical lens L S1, The position of the most concave portion b 1 is the same as that shown in FIG. 5 in the X axis direction. In that case, the position of the most convex part and / or the most concave part in the Z-axis direction is slightly adjusted. In this way, if you write the general formula, while maintaining the distance between the optical axes P CS2n-1 of the adjacent cylindrical lens constant as the distance P CS between the optical axes, the (2n-1) cylindrical lenses L S2n The cylindrical lens length P L2n-1 in the X axis direction of -1 can be different from the cylindrical lens length P L2n in the X axis direction of the second n cylindrical lens L S2n .

In the fourth method, the value of the radius of curvature R shown in FIG. 4 is changed to the (2n-1) th cylindrical lens L S2n-1 (for example, the first cylindrical lens L S1 ) and the second n cylindrical lens L S2n (for example, the first 2 cylindrical lenses L S2 ). Then, for example, the outermost convex portion a 1 of the first cylindrical lens L S1 of the outermost convex portion a 2 (optical axis) of the distance in the Z-axis direction from the image display surface 10 of the (optical axis) second cylindrical lens L S2 It was different from the slightly Figure 5 a distance in the Z axis direction from the image display surface 10, the same position when the position of the most recessed portion b 1 is shown in FIG. In this manner, it is possible to configure a stereoscopic image display optical element having two types of cylindrical lenses having different lengths and different curvatures.

Alternatively, the distance in the Z-axis direction from the image display surface 10 of the most convex portion a 1 (optical axis) of the first cylindrical lens L S1 and the image display of the most convex portion a 2 (optical axis) of the second cylindrical lens L S2. it may be made to a slightly different position of the most recessed portion b 1 and the distance in the Z-axis direction from the surface 10 as the same. That is, when the distance between the optical axis of the cylindrical lens and the optical axis of the adjacent cylindrical lens becomes short, the position of the most concave portion b 1 moves in the direction approaching the eyes of the viewer on the Z axis, and the cylindrical lens of when the distance between the optical axis and the optical axis of the adjacent cylindrical lenses is long to move away from the eyes of the viewer on the Z-axis position of the top recess b 1. In this manner, a stereoscopic image display optical element having two types of cylindrical lenses having different lengths having the same curvature can be configured.

  Furthermore, the position values in the Z-axis direction of both the most convex part and the most concave part may be slightly different from those in FIG. In this manner, a stereoscopic image display optical element having two types of cylindrical lenses having different lengths having the same curvature or different curvatures can be configured.

Be employed any more of the three as the fourth method, to write the general formula, while maintaining the distance between the optical axes P CS2n-1 of the adjacent cylindrical lenses constant to the inter-optical axis distance P CS The length (X-axis direction length) P L2n-1 of the (2n-1) th cylindrical lens L S2n-1 is different from the cylindrical lens length P L2n of the second n-th cylindrical lens L S2n in the X-axis direction. Can be made.

  The third method or the fourth method described above can be extended to a method for manufacturing a stereoscopic image display optical element having any kind of cylindrical lens. When many types of concave grooves are cut into a mold using the third method, the same number of cutting tools as the number of types of concave grooves are required, which increases the cost of mold manufacture. When many types of concave grooves are cut into the mold using the fourth method, the cost of mold manufacture can be reduced because only the position in the Z-axis direction is switched to be various types.

  FIG. 9 is a cross-sectional direction perpendicular to the image display surface (AA cross section in FIG. 8) showing the positional relationship between the image display surface of the stereoscopic image display device according to the second embodiment and a stereoscopic image display optical element having a plurality of cylindrical lenses. FIG.

In FIG. 9, the relationship between the pixels of the stereoscopic image display device and the cylindrical lens is schematically shown by a cross-sectional view. When FIG. 9 and FIG. 5 are compared, the dimensions relating to the image display surface 10 are the same. That is, the sub-pixel length P SPX in the X-axis direction of one sub-pixel is the same as the light-shielding part length P SX in the X-axis direction of one sub-pixel. In addition, the dimensions of the length P Pn for five subpixels are all the same length P P regardless of the value of n.

On the other hand, as described above, the dimensions related to the stereoscopic image display optical element 11B are generally set so that the lengths of the cylindrical lenses in the X-axis direction are different. In one specific example described above, every other cylindrical lens has a different length in the X-axis direction. Sum of the lengths of two different cylindrical lenses is twice the length P P.

When the separation distance between the stereoscopic image display optical element 11B and the straight line connecting the left eye and the right eye is in the vicinity of the distance L DZ1 , the nearest three subpixels of R, G, and B denoted by reference numeral 2 represent one pixel. It can be visually recognized by the right eye as configured, and can be visually recognized by the left eye so that the latest three sub-pixels of R, G, and B denoted by reference numeral 5 constitute one pixel. Further, when the separation distance is in the vicinity of the distance L DZ2 , the nearest three subpixels of R, G, and B with reference numeral 3 can be visually recognized by the right eye so as to form one pixel. The latest three subpixels of R, G, and B with 4 can be visually recognized by the left eye so as to form one pixel. The point that the image visually recognized by the right eye and the image visually recognized by the left eye are recognized as a stereoscopic image by the human is the same as the principle described with reference to FIG.

  In the case where the stereoscopic image display optical element 11B shown in FIGS. 8 and 9 is used, there are two types of axial lengths having the curvature of the cylindrical lens. Therefore, the regularity of interference of light from each cylindrical lens is disordered. Thus, the degree of occurrence of 3D moire is reduced. In addition, as shown in Equations 11 and 12, in general, the length of the cylindrical lens in the direction of curvature is not limited to two types, and any type of cylindrical lens can be used. The degree of occurrence of 3D moire is reduced as the number of types of cylindrical lenses increases.

(Third embodiment)
The third embodiment is an embodiment in which the types of cylindrical lenses in the first embodiment are increased, and the length of the cylindrical lens in the direction having the curvature is variously changed based on random numbers. The stereoscopic image display optical element used in the third embodiment can be realized by the first method or the second method already described.

  The length of the cylindrical lens constituting the stereoscopic image display optical element in the direction having the curvature may be a large number based on random numbers. Even in this case, as long as the above-described Expressions 3 and 4 are satisfied, the stereoscopic effect of the image is not impaired, and the quality of the image is not impaired. The stereoscopic image display optical element used in the third embodiment can be realized by the first method or the second method already described. The third embodiment corresponds to the case where a number of types of lengths in the direction having the curvature of the cylindrical lens (length in the X-axis direction) are selected in the stereoscopic image display optical element 11A shown in FIGS.

Also in the third embodiment, the equations 1 to 4 are applied as they are. The cylindrical lens length P Ln in the X-axis direction in Equation 1 is determined by a random number. The random number is, for example, a normally distributed random number. The characteristic of the cylindrical lens length P Ln in the X-axis direction that is normally distributed is represented by an average and a standard deviation value. The average of the cylindrical lens length P Ln in the X-axis direction is P P having a length of 5 subpixels. If deviations from the mean P P in the X-axis direction of the cylindrical lens length P Ln and delta n, the cylindrical lens length P Ln in the X-axis direction is represented by the number 13.

  Equation 14 is derived from Equation 1 and Equation 13.

Equation 15 is derived from Equation 3 and Equation 14. The meaning of Expression 15 is the same as that of Expression 3, and the above-described phase difference must be within the range indicated by Expression 15 in order to make the most concave portion b n within the range of the light shielding portion.

The meanings of Expressions 15 and 16 are the same as Expressions 3 and 4, and the distance difference (phase difference) between the distance L Cn and the distance L Pn must be within a range of one subpixel, The value defines an acceptable range of degradation of stereoscopic characteristics and image quality. According to the inventors' experiment, the value of k was less than 0.25. Equations (15) and (16) are mathematical expressions that determine how to generate random numbers. A specific random number generation method will be described later.

(Modification of the third embodiment)
Modification of the third embodiment, adjacent to the outermost convex portion a n of n-th cylindrical lenses constituting the stereoscopic image display optics (n + 1) -th distance between the maximum protrusion a n + 1 of the cylindrical lens (the distance between the optical axis) Is a large number based on random numbers. Even in this case, as long as the above-described Expressions 3 and 4 are satisfied, the stereoscopic effect of the image is not impaired, and the quality of the image is not impaired. That is, a modification of the third embodiment is to select a large number of optical axis distances by random numbers. The stereoscopic image display optical element used for the modification of the third embodiment can be realized by the first method or the second method already described. The modification of the third embodiment corresponds to the case where a large number of types of inter-optical axis distances P CSn are selected in the stereoscopic image display optical element 11A shown in FIGS.

In the modification of the third embodiment, the optical axis distance P CSn in Equation 1 is determined by a random number. The random number is, for example, a normally distributed random number. The characteristic of the inter-optical axis distance P CSn that is normally distributed is represented by an average and a standard deviation value. The average of the optical axis distance P CSn is set to P P having a length of 5 subpixels. (N-1) th of the cylindrical lens and the n-th cylindrical lens and of the deviation from the mean of the distance between the optical axes P CSn and delta n, inter-optical axis distance P CSn is represented by the number 17. Equations 18, 19, and 20 are obtained in the same manner as described above.

As shown in Equations 19 and 20, in the modification of the third embodiment, the inter-optical axis distance P CSn is determined by a random number. The degree of occurrence of 3D moire is reduced as the types of inter-optical axis distances PCSn increase. There is a relationship of sub-pixel length P SPX = P P / NSH, and according to experiments conducted by the inventors, the value of k was less than 0.25. Equations 19 and 20 are mathematical formulas that determine how to generate random numbers. A specific random number generation method will be described later.

(Fourth embodiment)
The fourth embodiment is an embodiment in which the types of cylindrical lenses in the second embodiment are increased. It may be a large number on the basis of different lengths P Ln of the n-th cylindrical lenses constituting the stereoscopic image display optical element in the random number. In this case, the image quality such as luminance unevenness can be maintained by satisfying the above-described Expression 7, and the stereoscopic effect of the image is not impaired as long as Expression 8 is satisfied. That is, in the fourth embodiment, the distance between the optical axes is kept constant while selecting a large number of cylindrical lenses by random numbers. The stereoscopic image display optical element used in the fourth embodiment can be realized by the third method or the fourth method already described. The fourth embodiment corresponds to the case where a large number of types of axial cylindrical lens lengths P Ln having curvature are selected in the stereoscopic image display optical element 11B shown in FIGS.

Also in the fourth embodiment, the equations 6 to 8 are applied as they are. The cylindrical lens length P Ln in the X-axis direction of the n-th cylindrical lens L Sn in Equation 6 is determined by a random number. The random number is, for example, a normally distributed random number. The average of the cylindrical lens length P Ln in the X-axis direction of the cylindrical lens L Sn is assumed to be P P having a length of 5 subpixels. Assuming that the deviation from the average of the cylindrical lens length P Ln in the X-axis direction of the cylindrical lens L Sn is Δn, the distance L Cn from the origin in the X-axis direction of the most concave portion b n of the cylindrical lens L Sn is expressed by Equation 21. Is done.

  On the other hand, the distance from the X-axis origin (X = 0) of the sub-pixels in units of five sub-pixels on the image display surface 10 is expressed by Equation 2. Therefore, Equations 22 and 23 are established.

The length of each subpixel in the X-axis direction is larger than the length of the light shielding portion in the X-axis direction, and P SPX > P SX . The features of Equations 22 and 23 are that the value of the random number that should be generated at present does not depend on the value of the past random number, that is, it is necessary to consider the value of the random number that appeared in the past and generate the current random number. As a result, random number processing is extremely simple. The value of k was less than 0.25 according to the experiment as described above.

In addition, the axial length (X-axis direction length) P Ln having the curvature of the n-th cylindrical lens L Sn is obtained by Equation 24.

(Regarding random numbers used in the third and fourth embodiments)
Third Embodiment In the fourth embodiment, how to set the deviation delta n from the length P L of the average of the cylindrical lenses will be described. Although the deviation delta n is set based on a random number, to be controlled a random number, it is impossible to hold the respective formula.

A random number used in the third embodiment will be described. Number 15, as a number 16 is clear, n-th ABS [L Pn -L Cn] is dependent on the deviation delta n based on all of the random numbers in the past. Therefore, special consideration is required for the generation of the deviation delta n of the third embodiment.

First, the standard deviation value 3σ is determined. Next, a random number having a standard deviation value 3σ is generated by a computer. Since a random number having a value exceeding the standard deviation value 3σ is theoretically generated, in that case, the value is replaced with the standard deviation value 3σ. Since the deviation delta n corresponds to the random number value delta n, in the following denoted the deviation delta n and random number delta n. A first random number Δ 1 , a second random number Δ 2 ,..., An nth random number Δ n is generated.

However, delta 1 and the optical axis distance P CS1 to correspond to variations in the P P, delta 2 and in correspondence with the variation in the P P of the distance between the optical axes P CS2, delta n of inter-optical axis distance P CSn as used when the number 15 the random numbers generated by a computer so as to correspond to variations in the P P, number 16, number 19, can not meet the number 20. This is because the random numbers generated by the computer are clearly unrelated to Equations 15, 16, 19, and 20. Therefore, the following process in addition to the random number delta n generated by a computer, the number 15, number 16, number 19, to generate a random number that satisfies Expression 20.

  In the following, a method for generating random numbers that satisfy Equations 15 and 16 will be described. The same method can be adopted for Equations 19 and 20.

The SUM shown in Equation 25 is obtained. SUM is the sum of up to n-th random number delta n having positive and negative polarities, the number 15 is the same as the number of 16. Since the positive / negative polarity is given by the calculation of Equation 26 described later, the generated random number may be a single polarity value.

After obtaining the absolute value of the result of addition from Δ 1 to Δ n in this way, −1 is generated when the SUM of Expression 25 is positive, and +1 is generated when the SUM of Expression 25 is negative. Then, as shown in Expression 26, and outputs the delta n dated polarity as variations in the distance between P CS optical axis.

Thus, the number 25, while applying polarity delta n as the result of adding the delta 1 to delta n by the calculation of the number 26 approaches zero, the number 15, the delta n satisfying the Expression 16 are sequentially generated, and assigns the delta n in number 13, can determine the cylindrical lens length in the X-axis direction having a curvature.

The number 22, for the occurrence of such delta n satisfy the number 23 will be described. The number 22, in number 23, because only a current value of delta n does not affect the formula, sequentially generates random numbers dated polarity from the beginning, it may be a value of the random number as a delta n.

(Other modifications of the first to fourth embodiments)
FIG. 10 is a diagram illustrating a stereoscopic image display optical element in which a cylindrical lens is arranged to be inclined in a direction that forms an angle θ with respect to the Y axis. The angle θ formed with respect to the Y axis of the cylindrical lens is θ = tan −1 (the length of the sub-pixel in the X-axis direction / the length of the sub-pixel in the Y-axis direction) = tan −1 (1/3) ≈18.43 Expressed in degrees. Then, similarly to the above, the stereoscopic image display optical element 11C that varies the length in the X-axis direction of each of the cylindrical lens L S1 , the cylindrical lens L S2 , the cylindrical lens L S3 , and the cylindrical lens L S4 is employed. 3D moiré is reduced.

  FIG. 11 is a diagram showing a stereoscopic image display optical element 11D that shifts the position of the cylindrical lens not only in the X-axis direction but also in the Y-axis direction. In this way, black lines due to 3D moiré vary in the X-axis direction, and the effect of reducing 3D moiré increases.

  FIG. 12 is a diagram showing another stereoscopic image display optical element 11E that shifts the position of the cylindrical lens not only in the X-axis direction but also in the Y-axis direction. The way of shifting is different from the stereoscopic image display optical element 11D of FIG. Even in this case, as in FIG. 11, the black line due to 3D moire is more varied in the X-axis direction than in FIG. 11, and the effect of reducing 3D moire is further increased.

  FIG. 13 is a diagram illustrating the stereoscopic image display optical element 11F in which the concave portion of the cylindrical lens smoothly swells. Unlike FIGS. 11 and 12, in this way, the mold can be easily manufactured, and the cylindrical lens can be easily injection-molded or pressure-molded. Even in this manner, as in FIGS. 11 and 12, the black lines due to 3D moire also vary in the X-axis direction, and the reduction effect of 3D moire is increased.

(Regarding the random number of the embodiment)
In 3rd Embodiment and 4th Embodiment, the random number was demonstrated as a normal distribution random number. Further, in the first embodiment and the second embodiment, if + δ and −δ are regarded as random numbers, the generated random number is a binary random number of + δ and −δ, and the variance is δ. It can be considered that the occurrence of -δ is regular. In general, even if the occurrence of + δ and −δ is random, it can be performed without losing generality. Therefore, the first embodiment is included in the third embodiment, and the second embodiment is the fourth embodiment. It can be said that it is included in the form.

  In short, as explained in the principle of the embodiment, in order to reduce 3D moire, it is only necessary to disrupt the arrangement of the cylindrical lenses with uniformity. It may be a uniform random number, a triangular distribution random number, an exponential distribution random number, or the like. In the third and fourth embodiments, random numbers in the range of the standard deviation value 3σ of normally distributed random numbers are used. However, random numbers in the range of standard deviation value 2σ and random numbers in the range of standard deviation value σ are used. You may do it. When the standard deviation value is the standard deviation value 3σ, the same polarity continuity increases, the standard deviation value 2σ, the standard deviation value σ decreases in the order of the same polarity, and the standard deviation value σ Positive polarity and negative polarity can be generated almost alternately.

(For limit Polarized deviation delta n)
The maximum value of the polarity with the deviation delta n for (limit not increase any more deviation) will be described. Although must be subtracted the variation as a margin at the time of manufacture, the maximum value of Polarized deviation delta n is the first embodiment, second embodiment, third embodiment, fourth embodiment, as described above in a few It is determined so as to satisfy the formulas 15, 15, 19, 20, 20, and 23. Here, it is determined under the condition that the value of k in Equations 16, 20, and 23 is less than 0.25.

The minimum value of the polarity with the deviation delta n for (limit not the maximum value of the deviation delta n more small) will be described. If the minimum value of the deviation Δn with polarity is 0, the principle of the embodiment is the same. Minimum value of Polarized deviation delta n is Sadamari in view of the reduction of 3D moire, and be determined as experimental value. According to the inventor's experimental results, when the value of k in Equation 16, Equation 20, and Equation 23 is 0.014 (1.4%), the probability that Equation 16, Equation 20, and Equation 23 do not hold. sigma, or 2 [sigma], or the minimum value of the deviation delta n such that 3σ is determined.

The relationship between the third embodiment and the fourth embodiment will be described. Expressions 15, 16, 19, and 20 in the third embodiment are compared with Expressions 22 and 23 in the fourth embodiment. In Equations 15, 16, 19, and 20, it seems that the influence of the first to nth variations affects the nth variation. However, when performing calculations of Formula 15, Formula 16, Formula 19, and Formula 20, calculations of Formula 25 and Formula 26 are performed, and Δ 1 in Formula 15, Formula 16, Formula 19, Formula 20, and so on. -Both [Delta] n-1 and [Delta] n are polar random numbers. This is because, as already described, if this is not done, the phase error accumulates as the distance from the origin increases. Here, the random numbers with polarity are always managed so as to approach 0. That is, the random numbers in the formulas 15, 16, 19, and 20 are all the random numbers from the first to the n-th, and the random numbers in the formulas 22 and 23 are the axes having the curvature of the cylindrical lens. Although it appears in mathematical formulas to affect the length variation in the direction, there is no change in that the length variation in the axial direction with the curvature of each cylindrical lens is controlled by random numbers of the same nature .

  The purpose of the third embodiment is to vary the length of the cylindrical lens and / or the distance between the optical axes, and the purpose of the fourth embodiment is to vary the length of the cylindrical lens and reduce the distance between the optical axes. As a result, in the third embodiment and the fourth embodiment, as a result, the length of the cylindrical lens is varied, so that it is between the third embodiment and the fourth embodiment. There is no essential difference. From the viewpoint of the optical effect produced by the manufactured stereoscopic image display optical element, the stereoscopic characteristic deterioration in the fourth embodiment is the first in that the distance between the optical axes is kept unchanged in the fourth embodiment. Smaller than in the third embodiment.

(Fifth embodiment)
Although the fifth embodiment shares the technical idea with the first to fourth embodiments described above in that the uniformity of the cylindrical lens is lost, the correspondence relationship between the cylindrical lens and the sub-pixel is different.

  In the first to fourth embodiments, the viewer can visually recognize one pixel by the R, G, and B subpixels arranged in order in the same direction as the axial direction having the curvature of the cylindrical lens. (See FIGS. 6 to 13). However, in the fifth embodiment, R, G, and B subpixels are arranged in a direction that does not have the curvature of the cylindrical lens, and subpixels that emit light of the same color are arranged in the direction that has the curvature of the cylindrical lens. The The viewer can visually recognize the pixel as 1 pixel by light from the R, G, and B sub-pixels incident on the viewer's eyes from the same direction. That is, in the fifth embodiment, the relative relationship between the arrangement of the cylindrical lenses and the arrangement of the sub-pixels is 90 ° different from that of the first to fourth embodiments. Even in such an arrangement, the general theory described in the first to fourth embodiments holds. The fifth embodiment will be described with reference to FIG.

  FIG. 14 is a diagram showing a stereoscopic image display optical element 11G that faces two sub-pixels in the axial direction (Y-axis direction) in which the image display surface 10 is inclined by 90 ° and each cylindrical lens has a curvature. In FIG. 14, an example of the two-parallax method will be described. However, the following description is not limited to the two-parallax method, and applies to other parallax methods. For example, when each cylindrical lens faces four sub-pixels in the axial direction (Y-axis direction) having a curvature, the four-parallax method is used, but the following description also applies to the four-parallax method.

  In the fifth embodiment, an example of the two-parallax method will be described. However, since the same image display surface 10 as in the first to fourth embodiments is used, the coordinate axes of the X axis and the Y axis are the image display surface 10. Is based on. In the fifth embodiment, unlike the first to fourth embodiments, each cylindrical lens of the stereoscopic image display optical element 11G has a curvature in the Y-axis direction. Each cylindrical lens is arranged in the Y-axis direction and faces two sub-pixels that emit light of the same color, so that a stereoscopic image can be visually recognized by the two-parallax method. Each of the right-eye pixel and the left-eye pixel is indicated by a broken line. The other sub-pixels are similarly configured as right-eye pixels and left-eye pixels.

  FIG. 15 is a sectional view perpendicular to the image display surface (BB in FIG. 14) showing the positional relationship between the image display surface 10 of the stereoscopic image display device according to the fifth embodiment and the stereoscopic image display optical element 11G having a plurality of cylindrical lenses. FIG. 14, R (red) is replaced with G (green) in FIG. 15 in the CC sectional view of FIG. 14, and R (red) is replaced with B (blue) in FIG. 15 in the DD sectional view of FIG. 14. Then, as shown in FIG. 15, the viewer recognizes R, G, and B (inside the broken line in FIG. 14) aligned in the Y-axis direction, which is the axial direction having no cylindrical lens curvature, as one pixel.

In the fifth embodiment, as in the third embodiment, the distance between the optical axes of the cylindrical lenses (see the reference characters L S1 , L S2 , L S3 , and L S4 in FIGS. 14 and 15) (FIG. 14). The code P CS1 , the code P CS2 , and the code P CS3 in FIG. 15) may be varied. Further, in the fifth embodiment, similarly to the fourth embodiment, the cylindrical lenses (see the signs L S1 , L S2 , L S3 , and L S4 in FIGS. 14 and 15) have a curvature direction. The lengths (see symbols P L1 , P L2 , P L3 , and P L4 in FIGS. 14 and 15) may be varied.

  Expressions 27 and 28 correspond to Expressions 15 and 16.

Expressions 29 and 30 correspond to Expressions 19 and 20.

  Expressions 31 and 32 correspond to Expressions 22 and 23.

Equations 27, 29, and 31 are conditional expressions for the most concave portion b n to be positioned within the range of the light shielding portion length P SY. Equations 28, 30, and 32 are the most concave portion b n is k × 1 subpixels or more. This is a conditional expression for preventing deviation. k is a constant, and the value of k was less than 0.25 according to experiments. Further, P SPY = P P / NSH. Here, PP is the length of two R (green) sub-pixels denoted by reference numeral 1 in FIG.

(Measured data of the example)
The inventors made a prototype of a stereoscopic image display optical element that realizes the two-parallax method shown in FIGS. The cylindrical lens length of this stereoscopic image display optical element is set using random numbers based on Equations 27 and 28. Table 1 is a table summarizing the measured values of the cylindrical lens length of the prototype stereoscopic image display optical element. The measurement was performed using a contact-type measuring instrument equipped with a 2 μm diamond needle.

The maximum value of the cylindrical lens length was 0.157 mm. The minimum value of the cylindrical lens length was 0.15475 mm. The average cylindrical lens length was 0.1559 mm. The reference length P P of the cylindrical lens length in the design is 0.156 mm.

Standard deviation σ of the deviation delta n is 0.000487798mm = 0.487798μm, includes 68.3% of the total deviation. The standard deviation value 3σ is 0.00146363393 mm = 1.463393 μm, and 99.7% of the entire deviation is included.

  FIG. 16 is a diagram illustrating a part of actual measurement data indicating a deviation from the average of the cylindrical lens lengths of the cylindrical lenses of the stereoscopic image display optical element according to the prototype.

The horizontal axis of FIG. 16 is n, and the vertical axis represents the deviation delta n from the average of the cylindrical lens length. The unit of the scale on the vertical axis is mm, and the scale is engraved every 0.0005 mm = 0.5 μm (micrometer).

(Measurement data of comparative example)
The inventor made a prototype of a stereoscopic image display optical element close to the condition of Formula 5 as a comparative example. That is, the cylindrical lens length of the stereoscopic image display optical element is set to be a constant value, but some variation has occurred in the cylindrical lens length due to an error in manufacturing. Table 2 is a table summarizing the measured values of the cylindrical lens length of the stereoscopic image display optical element of the comparative example that was experimentally manufactured. The measurement was performed using a contact-type measuring instrument equipped with a 2 μm diamond needle.

  The maximum value of the cylindrical lens length was 0.15675 mm. The minimum value of the cylindrical lens length was 0.15475 mm. The average cylindrical lens length was 0.15600556 mm. The standard deviation value σ is 0.00034377 mm = 0.34377 μm, and 68.3% of the total deviation is included. The standard deviation value 3σ is 0.0010313 mm = 1.0313 μm, and 99.7% of the entire deviation is included.

  FIG. 17 is a diagram illustrating a part of actual measurement data indicating a deviation from the average of the cylindrical lens lengths of the cylindrical lenses of the prototype stereoscopic image display optical element of the prototype.

The horizontal axis of FIG. 17 is n, and the vertical axis represents the deviation delta n from the average of the cylindrical lens length. The unit of the scale on the vertical axis is mm, and the scale is engraved every 0.0005 mm = 0.5 μm (micrometer).

(Comparison of actual measurement data between Example and Comparative Example)
Table 1 and Table 2 are compared.

(1) Contrast standard deviation value σ. It is 0.00034377 mm (comparative example) with respect to 0.0004877798 mm (example).
(2) Contrast standard deviation value 3σ. It is 0.0010313 mm (comparative example) with respect to 0.001463393 mm (example).
(3) Both the standard deviation value σ and the standard deviation value 3σ are larger in the example than in the comparative example. In the examples, random lens lengths are used to positively vary the cylindrical lens length, and in the comparative examples, the cylindrical lens length is positively managed to a constant value.
(4) As a conclusion, the stereoscopic image display optical element of the comparative example is formed with a more uniform cylindrical lens than the stereoscopic image display optical element of the example.

  FIG. 16 and FIG. 17 are compared.

There exists the following tendency about an Example (refer FIG. 16).
(1) deviation delta n embodiments, larger deviation delta n of the comparative example.
(2) polarity of the deviation delta n of the cylindrical lens length is relatively, are distributed alternately positive and negative. This is a phenomenon caused by manufacturing a stereoscopic image display optical element by adopting a method in which the cylindrical lens length is varied by a random number while managing the added value from the origin of the most concave portion of the cylindrical lens in the X-axis direction. It is.

The comparative example (see FIG. 17) has the following tendency.
(1) Deviation delta n of the comparative example is smaller than the deviation delta n embodiment.
(2) polarity of the deviation delta n of the cylindrical lens length is relatively positive continuous, often negative continuous. This is a phenomenon that occurs because the method of constantly managing the cylindrical lens length in the X-axis direction is employed.

(Contrast of 3D moire between Example and Comparative Example)
FIG. 18 is a diagram comparing 3D moire between the example and the comparative example.

  The same display was performed on each image display surface of two smartphones of the same model number, and the comparison was made with the same brightness and the same conditions. FIG. 18A is a photograph in which the stereoscopic image display optical element of the example is mounted on a smartphone. FIG. 18B is a photograph in which the stereoscopic image display optical element of the comparative example is attached to a smartphone.

  In the embodiment of FIG. 18A, the 3D moire is thin and inconspicuous. In the example of FIG. 18B, the shade of 3D moire is dark and conspicuous. As can be seen from FIG. 18, the amount of 3D moire can be reduced by using a stereoscopic image display optical element in which the cylindrical lens length is positively varied.

  The physical dimensions of the smartphone (flat image display device) used in the examples and comparative examples will be described. The smartphones of the examples and comparative examples have the same dimensions and the same characteristics.

The size of the subpixel is as follows. The sub pixel length P SPX (including the light shielding portion length P SX ) in the X axis direction of one sub pixel is 0.026 mm. The light shielding part length P SX is 0.00085 mm. In addition, the sub-pixel length P SPY (including the light shielding portion length P SY ) in the Y-axis direction of one sub-pixel is 0.078 mm (millimeter). Shielding director P SY is 0.00995Mm.

In an embodiment, the standard deviation values 3σ that contains 99.7% of the total of all the cylindrical lens length variation amount is 0.001463393Mm, less than 0.00995mm light shielding director P SY. As a result of the comparative evaluation of the stereoscopic image quality other than the 3D moire between the example and the comparative example, no difference was observed in the stereoscopic characteristics and the image quality in the visual experiment of the stereoscopic image.

This is because the concave portions b 1 , b 2 ,... Of the cylindrical lens are arranged within a range of 0.00995 mm of the light shielding portion length P SY , so that the concave portions of each cylindrical lens having a great influence on the image quality. This is because does not face the light emitting portion of the subpixel. Further, k = (standard deviation value 3σ) / (sub-pixel length P SPY in the Y-axis direction of the sub-pixel) = 0.00146363393 mm / 0.078 mm≈0.0188, and the value of k is compared with 0.25. Because it is much smaller, it has little effect on the stereoscopic characteristics.

(Contrast of spatial frequency between Example and Comparative Example)
FIG. 19 is a diagram illustrating a comparison between the spatial frequency of the stereoscopic image display optical element of the example and the spatial frequency of the stereoscopic image display optical element of the comparative example obtained by calculation.

  In the calculation of the spatial frequency, in each of the example and the comparative example, data on the distance in the Z-axis direction with respect to the distance in the axial direction (Y-axis direction in FIG. 14) having the curvature of the cylindrical lens of the actually measured stereoscopic image display optical element is obtained. Each of the data was taken and compared by Fourier transform (FFT). The vertical axis in FIG. 19 is power, and the horizontal axis is spatial frequency.

  In the comparative example, since the cylindrical lenses are regularly arranged, only the fundamental frequency and the main spatial frequency components are visible. On the other hand, in the embodiment, a sub-peak is generated by having a cylindrical lens having a different lens length with respect to the average value of the axial lens length having a curvature, and the dispersion of the peak due to the average lens length is also widened. ing. The fact that the spatial frequency is distributed over a wide range means that the periodicity necessary for generating moire is lost and 3D moire is hardly generated.

(Deviation Δ n setting)
In an example, the number 27, for it has how to set the deviation delta n in Equation 28 will be described.

First, a method for determining the standard deviation value 3σ will be examined. The value of the standard deviation values 3σ is the length of the light shielding director P SY, it is desirable to less 0.00995Mm. Since 99.7% of the total random number is included in this deviation, the position of the most concave portion of each cylindrical lens manufactured based on the random number hardly faces the light emitting portion of the subpixel, and the image quality There will be almost no deterioration of the three-dimensional effect. Further, desirably, when a random number having a value exceeding the standard deviation value 3σ is generated, the value is replaced with the standard deviation value 3σ. That is, the random number according to the normal distribution theoretically generates a large deviation with a small probability, and is therefore limited to the standard deviation value 3σ by the limiter. Even in this way, random numbers that can be almost regarded as a normal distribution can be obtained. In the embodiment, as shown in Table 1, including manufacturing variations, the standard deviation 3σ is 0.001463393Mm, the length of the light shielding director P SY, is smaller than the 0.00995Mm.

  It should be noted that although the 3D moire is reduced as the variation in the length of the cylindrical lens in the direction of curvature naturally increases as the stereoscopic image quality decreases, the stereoscopic image quality deteriorates. There is a problem. That is, the problem is how to determine the value of k in Equation 4, Equation 9, Equation 12, Equation 16, Equation 20, Equation 23, Equation 28, Equation 30, and Equation 32.

The inventor gradually shifted the position of the stereoscopic image display optical element 11G in the Y-axis direction with respect to the image display surface 10 to measure the phase shift between the two at a point recognized by the viewer when the stereoscopic characteristics and image quality deteriorate. . As a result, no significant deterioration in stereoscopic characteristics and image quality was recognized up to 1/4 = 0.25 of the sub-pixel length P SPY in the Y-axis direction of the sub-pixel. That is, even if the position of the most concave portion of the cylindrical lens of the stereoscopic image display optical element 11G is shifted within a range of less than ¼ (0.25) with respect to the position of the subpixel boundary, the stereoscopic characteristics and the image quality are deteriorated. There will be almost no. The inventor has confirmed that the value of 1/4 (0.25) (value of k) is substantially constant regardless of the parallax method such as the 2-parallax method, the 5-parallax method, and the subpixel structure.

  Also, what is the minimum variation value that can exhibit the 3D moire reduction effect is a problem. Regarding this point, the comparative example has almost no effect of reducing the 3D moire, and based on the experimental fact that the effect of reducing the 3D moire was sufficient in the example, it is the minimum that can exhibit the effect of reducing the 3D moire. The limit of variation was set.

Regarding the embodiment, the amount of variation in the cylindrical lens length from Table 1 is 0.157 mm (maximum) −0.156 mm (design value) = + 0.001 mm. Further, 0.15475 mm (design value) −0.156 mm (design value) = − 0.00125 mm. The average of absolute values on the + side and − side is 0.001125 mm. In the two-parallax method in the embodiment, as one design index of the minimum variation amount of the cylindrical lens for reducing 3D moire, 0.001125 mm (average value of absolute values on the + side and − side) /0.078 mm ( The inventor's experiment shows that if the subpixel length P SPY in the Y-axis direction of the subpixel is approximately 0.014, there is a moire reduction effect as shown in FIG.

Regarding the comparative example, from Table 2, the amount of variation in the cylindrical lens length is 0.15675 mm (maximum) −0.156 mm (design value) = + 0.00075 mm. Further, 0.15475 mm (minimum) −0.156 mm (design value) = − 0.00125 mm. The average absolute value of the + side and the − side is 0.001 mm. In the comparative example, when the same index as in the example is employed, 0.001 mm (average value of absolute values on the + side and − side) /0.078 mm (sub pixel length P SPY in the Y axis direction of the sub pixel) ≈ The inventor's experiment shows that if it is 0.012, there is almost no moire reduction effect as shown in FIG.

  From the above examples and comparative examples, in general, the average value (reference length) of the axial length having the curvature of the cylindrical lens (reference length) is (0.001125 / (0.156 / NSH), that is, (reference length Variation of (a cylindrical lens length / predetermined number according to the parallax method) = 0.014, if it is varied, an effect of reducing 3D moire has occurred, whereas an average value of axial lengths having the curvature of the cylindrical lens ( (Reference length) (0.001 / (0.156 / NSH), that is, (variation from reference length / (cylindrical lens length / predetermined number according to parallax method)) = 0.012 (The cylindrical lens length / a predetermined number according to the parallax method) is the length of the sub-pixel in the axial direction having the curvature of the cylindrical lens. It. In the case of 2 parallax method NSH is 2.

  From this experimental result, the maximum value of the variation causing the 3D moire reduction effect is 0.028 / NSH or more of the average value (reference length) of the axial length having the curvature of the cylindrical lens. In terms of the value of k, the minimum value of k that produces the 3D moire reduction effect is 0.014.

(Key points of variation of cylindrical lenses that reduce 3D moire)
The main points of the variation of the cylindrical lens that reduces the 3D moire common to all the embodiments will be described based on the experimental results.

  The greater the variation in the length of the cylindrical lens in the direction of curvature, the greater the 3D moire reduction effect, but the greater the degradation of the stereoscopic image quality. As described above, even if there is a variation in which the value of k is less than 0.25, that is, the length of the subpixel in the direction having the curvature of the cylindrical lens is less than ¼ = 0.25, the image quality of the stereoscopic image is not deteriorated. Almost inconspicuous. On the other hand, if the value of k is 0.014 or more, a 3D moire reduction effect is produced. Therefore, the value of k is preferably 0.014 (1.4%) or more and less than 0.25 (25%). Further, from the viewpoint of 3D moire reduction effect and stereoscopic image quality, the value of k is more than 0.028 (2.8%) and less than 0.125 (12.5%). desirable.

  Furthermore, it is more desirable that the value of k is less than (the length of the light shielding portion in the direction having the curvature of the cylindrical lens) / (the length of the subpixel in the direction having the curvature of the cylindrical lens). In such a range, the most concave portion of the cylindrical lens can be prevented from covering the light emitting portion of the sub-pixel, so that the image quality can be hardly deteriorated.

  Since the value of k described above is determined on the basis of the length of the sub-pixel, it holds regardless of the parallax method. On the other hand, when the length of the cylindrical lens is used as a reference, different expressions can be made according to the parallax method.

Assuming that the number of subpixels arranged in the direction having the curvature of the cylindrical lens is a predetermined number NSH, the relationship between the subpixel length and the cylindrical length is expressed by Expressions 33 and 34. Equation 33 is a case where the cylindrical lens has a curvature in the X-axis direction, and Equation 34 is a case where the cylindrical lens has a curvature in the Y-axis direction. NSH is 2 for the 2 parallax method, 4 for the 4 parallax method, 5 for the 5 parallax method, and 7 for the 7 parallax method. P P is the sub-pixel number × 1 sub pixel length corresponding to the parallax method, referred to as a reference length of the cylindrical lens P P.

  The length of each cylindrical lens varies around the reference length of the cylindrical lens, and even if the length of the cylindrical lens is the same, the reference length of the cylindrical lens is the center depending on the parallax method used for the stereoscopic image display optical element. The allowable range of variation is different.

A description will be given of how to define an allowable amount of variation as a single stereoscopic image display optical element. Assuming that ks is the amount of variation allowed with respect to the cylindrical lens length (reference length) of an ideal cylindrical lens having perfect uniformity in which all cylindrical lens dimensions are P P , The k × P SX or k × P SY of the equations of Equation 16, Equation 20, Equation 23, Equation 28, and Equation 30 can be rewritten as ks × P p . Here, ks = k / NSH.

  Therefore, the maximum value of the variation of the cylindrical lenses sequentially arranged in the predetermined axis direction having the curvature of the stereoscopic image display optical element is (2.8% of the reference length which is the cylindrical lens length of the ideal cylindrical lens). / NHS) or more and less than (50% / NHS), that is, the value of ks is preferably 2.8% or more and less than 50%. That is, the maximum value of the difference between the position of the ideal cylindrical lens boundary where each cylindrical lens is uniform and the position of the boundary of each cylindrical lens of the embodiment is (2.8% / NHS) or more ( It is desirable that the range be less than 50% / NHS). Furthermore, from the viewpoint of 3D moire reduction effect and stereoscopic image quality, the value of ks is more preferably 0.056 (5.6%) or more and less than 0.25 (25%).

  A specific example of the design of a cylindrical lens having the parameters shown in Table 1 will be described below.

A random number having a standard deviation value 3σ of, for example, 0.0015 mm is generated by a computer. 1 first random number delta, the second random number delta 2, the · · · n th random number delta n is generated. However, Δ 1 is made to correspond to the variation of the cylindrical lens length P L1 from P P , Δ 2 is made to correspond to the variation of the cylindrical lens length P L2 from P P , and Δ n is changed from the P P of the cylindrical lens length P Ln. If the random numbers generated by the computer are used as they are so as to correspond to the variations in the above, the equations 27 and 28 cannot be satisfied. Therefore, the number on the random number delta n generated by computer 25, to perform the process several 26.

The minimum value of the polarity with the deviation delta n for (limit not reduced any more deviation), has been described that the 0.028 / NSH than is desired, as another indicator, the description below for the case of the standard deviation as an index To do.

In the example, the measured value of the standard deviation value 3σ of the variation (deviation Δ n ) of the cylindrical lens was 0.00146363393 mm. In the comparative example, the deviation = 0 was aimed, but the measured value of the standard deviation value 3σ after manufacturing the stereoscopic image display optical element was 0.0010313 mm. In the example and the comparative example, the length P P in the X-axis direction of the target 5 sub-pixels is 0.156 mm. In the example, (standard deviation value 3σ) / (cylindrical lens length P P ) = 0.146463393 mm / 0.156 mm≈0.0094. That is, (standard deviation value 3σ) ≈0.0094 × (cylindrical lens length P P ). On the other hand, in the comparative example, (standard deviation value 3σ) / (cylindrical lens length P P ) = 0.146463393 mm / 0.156 mm≈0.0093. That is, (standard deviation value 3σ) ≈0.0093 × (cylindrical lens length P P ).

From the above results, the following can be said at least under the parameters of the example and the comparative example. When (standard deviation value 3σ) ≈0.0093 × (cylindrical lens length P P ), the effect of improving 3D moire is small, but when (standard deviation value 3σ) ≈0.0094 × (cylindrical lens length P P ), 3D The improvement effect of moire is great.

(Manufacturing method of stereoscopic image display optical element with varying cylindrical lens length)
The parameters of each part of the stereoscopic image display optical element having a varying cylindrical lens length can be determined by the first method or the second method described above. After that, after molding a mold having this dimension, a technique for manufacturing a mold to be fitted with the mold, or a technique using an NC lathe, a laser processing apparatus, or a tool, a groove is formed on a nickel-plated flat plate. It is a well-known technique to manufacture a mold by a technique of digging a groove into a cylindrical shape that has been plated with copper by using a cutting tool. A stereoscopic image display optical element can be injection-molded or pressure-molded using such a mold. Such a molding technique is a well-known technique and is also described in Patent Document 5 cited in the Background Art section.

(Driving circuit unit of stereoscopic image display device)
A circuit unit for driving the image display surface 10 in the stereoscopic image display apparatus according to the above-described embodiment will be briefly described.

  FIG. 20 is a block diagram illustrating the entire stereoscopic image display apparatus according to the embodiment.

  A stereoscopic image display optical element 11A or the like (a stereoscopic image display optical element 11A or a stereoscopic image display optical element is arranged so as to face the image display surface 10 of the stereoscopic image display device 8 shown in FIG. 11G) is arranged.

  The circuit unit of the stereoscopic image display device 8 includes a control unit 20, a receiving unit 30, a Y-axis direction driving unit 40, and an X-axis direction driving unit 50. The Y-axis direction drive unit 40 and the X-axis direction drive unit 50 drive each of the subpixels constituting the image display surface 10. Instead of drawing out the wiring from each subpixel, a matrix wiring system is employed in which the luminance of the subpixel is controlled at the intersection of two-dimensional wiring in the X-axis direction and the Y-axis direction. As the driving method, either a simple matrix driving method which is a well-known technique or an active matrix driving method which is a well-known technique can be used in the embodiment, but the case of the active matrix driving method will be described below.

  In the active matrix driving method, a source line connected to the X-axis direction driving unit 50, a gate line (not shown) connected to the Y-axis direction driving unit 40, a storage capacitor (not shown), and an active element ( (Not shown) is provided for each sub-pixel. In general, a thin film transistor is used as an active element. By controlling the thin film transistor with the source line and the gate line, the sub-pixel connected to the thin film transistor whose gate line is selected simultaneously stores the voltage applied to the source line in the storage capacitor, and stores it when not selected The sub-pixel emits light with a luminance corresponding to the voltage stored in the capacitor.

  The receiving unit 30 includes a tuner 301 and an image signal detector 302. The tuner 301 selects a desired radio wave modulated by the stereoscopic image signal. The image signal detector 302 demodulates the stereoscopic image signal. The demodulated stereoscopic image signal is sent to the control unit 20.

  The control unit 20 includes a central processing unit (CPU) 201, a ROM (ROM) 202, a ram (RAM) 203, a left-eye right-eye subpixel signal generator 204, an image input interface 205, a Y-axis direction drive signal generator 206, An X-axis direction drive signal generator 207 is provided, and these units are connected by a bus line. The central processing unit (CPU) 201 is connected to a screen layout manual switching unit 208 and a screen layout automatic detector 209.

  The center of control in the control unit 20 is the central processing unit 201, and based on the program information stored in the ROM 202, other parts of the control unit 20 using the ram 203 functioning as a temporary storage unit of information, the receiving unit 30, the Y-axis direction drive unit 40 and the X-axis direction drive unit 50 are controlled to perform the following control so that a stereoscopic image is visually recognized on the image display surface 10.

  The central processing unit 201 controls the image input interface 205 to select either a stereoscopic image signal from a host device (not shown) or a stereoscopic image signal from the receiving unit 30. The stereoscopic image signal may be a still image or a moving image. The central processing unit 201 inputs a signal from the screen layout automatic detector 209 or the screen layout manual switcher 208 and determines whether to display a horizontally long screen layout or a vertically long screen layout on the image display surface 10. The central processing unit 201 controls the Y-axis direction drive signal generator 206 and the X-axis direction drive signal generator 207, and is driven by each subpixel signal obtained by the left-eye and right-eye subpixel signal generator 204. Is controlled on the image display surface 10.

  The Y-axis direction drive signal generator 206 controls the Y-axis direction drive unit 40, and the X-axis direction drive signal generator 207 controls the X-axis direction drive unit 50. Then, each subpixel is caused to emit light so that a desired stereoscopic image can be visually recognized by the source line and the gate line.

  Embodiments combining some or all of the above-described embodiments can also be implemented. Moreover, it goes without saying that the present invention is not limited to the above-described embodiments, but covers the scope of the same technical idea.

8 stereoscopic image display apparatus, 10 an image display surface, 11,11A, 11B, 11C, 11D , 11E, 11F, 11G stereoscopic image display optics, a 1, a 2, a 3, a n Saitotsu unit, b 0, b 1 , b 2 concave portion, L S1 , L S2 , L S3 , L S4 cylindrical lens, P CS , P CS1 , P CS2 , P CS3 optical axis distance, P L1 , P L2 , P L3 cylindrical lens length, P P1 , P P2 , P P3 , P Pn 5 subpixel length (length of 5 subpixels), P SPX length of 1 subpixel in the X-axis direction, P SPY 1 Y of subpixel Axial length, P SX (X-axis direction) light shielding part length, P SY (Y-axis direction) light shielding part length

Claims (5)

  1. A planar image display device regularly arranging a plurality of subpixels in a predetermined axis direction of the image display surface and in a direction orthogonal to the predetermined axis ;
    And a stereoscopic image display optical element a plurality of cylindrical lenses are formed by sequentially arranged in the predetermined axial direction extending in a direction have a curvature perpendicular to the predetermined axis in the predetermined axial direction,
    Each of the sub-pixels is
    A light emitting unit that emits light;
    A light-shielding part that does not emit light at an end of the light-emitting part in the predetermined axis direction,
    Forming a plurality of a predetermined number of subpixels composed of a predetermined number of the subpixels sequentially adjacent in the predetermined axis direction;
    Each of the cylindrical lenses,
    Facing each of the predetermined number of sub-pixels;
    The cylindrical lens has an irregular variation around a predetermined reference length in which the length in the predetermined axis direction of each of the cylindrical lenses is the length of the predetermined number of subpixels ,
    The maximum value of the irregular variation is
    For all of the above cylindrical lenses,
    The phase difference between the boundary of the cylindrical lens and the boundary of the predetermined number of subpixels is less than 25% of the length of the subpixel in the predetermined axis direction ;
    The stochastic minimum value of the irregular variation magnitude for all of the cylindrical lenses is:
    (Standard deviation 3σ of the irregular variation / predetermined reference length) is 0.0094 or more,
    Stereoscopic image display device.
  2. A planar image display device regularly arranging a plurality of subpixels in a predetermined axis direction of the image display surface and in a direction orthogonal to the predetermined axis ;
    And a stereoscopic image display optical element a plurality of cylindrical lenses are formed by sequentially arranged in the predetermined axial direction extending in a direction have a curvature perpendicular to the predetermined axis in the predetermined axial direction,
    Each of the sub-pixels is
    A light emitting unit that emits light;
    A light-shielding part that does not emit light at an end of the light-emitting part in the predetermined axis direction,
    Forming a plurality of a predetermined number of subpixels composed of a predetermined number of the subpixels sequentially adjacent in the predetermined axis direction;
    Each of the cylindrical lenses is
    Facing each of the predetermined number of sub-pixels;
    The cylindrical lens has an irregular variation around a predetermined reference length in which the length in the predetermined axis direction of each of the cylindrical lenses is the length of the predetermined number of subpixels ,
    The maximum value of the irregular variation is
    For all of the above cylindrical lenses,
    To face the boundary of the sheet Lind helical lens within the range of the predetermined number subpixel edge shielding portion is a light shielding portion of the end of the predetermined number subpixels,
    The phase difference between the boundary of the cylindrical lens and the predetermined number of subpixel boundaries is less than or equal to the length of the light shielding portion of the subpixel.
    The stochastic minimum value of the irregular variation magnitude for all of the cylindrical lenses is:
    (Standard deviation 3σ of the irregular variation / predetermined reference length) is 0.0094 or more,
    Stereoscopic image display device.
  3. The irregular variation is determined based on a random number,
    The variation of the nth cylindrical lens corresponding to an arbitrary positive integer n with the end in the predetermined axis direction of the stereoscopic image display optical element as a base point is
    Determined by the first random number to the nth random number,
    When the added value of the first random number to the nth random number is a positive value, the absolute value of the nth random number multiplied by −1 is the nth random number and the nth cylindrical lens. If the added value is a negative value, the absolute value of the nth random number multiplied by 1 is the variation length of the nth random number and the nth cylindrical lens. The
    The stereoscopic image display apparatus according to claim 1 or 2 .
  4. The irregular variation in the length of the cylindrical lens in the predetermined axial direction is:
    It is obtained by changing the radius of curvature of the cylindrical lens, the distance from the image display surface of the most convex portion of the cylindrical lens, or the distance from the image display surface of the concave portion of the cylindrical lens.
    The three-dimensional image display apparatus according to claim 1.
  5. A stereoscopic image display optical element mounted on a flat image display device in which subpixels are regularly arranged on an image display surface,
    Are formed by sequentially arranging a plurality of cylindrical lenses extending in a direction orthogonal to have a curvature the predetermined axis in a predetermined direction to the predetermined direction,
    The length of each cylindrical lens in the predetermined axial direction has irregular variations around a predetermined reference length that is the length of a predetermined number of the sub-pixels ,
    The maximum value of the irregular variation is
    For all of the above cylindrical lenses,
    The difference between the length from the end in the predetermined axial direction of the stereoscopic image display optical element to the boundary of the nth cylindrical lens corresponding to an arbitrary positive integer n and the n × predetermined reference length is the predetermined sub-pixel length. Less than 25% of the axial length,
    The stochastic minimum value of the irregular variation magnitude for all of the cylindrical lenses is:
    (Standard deviation 3σ of the irregular variation / predetermined reference length) is 0.0094 or more,
    Stereoscopic image display optical element.
JP2013054468A 2013-03-17 2013-03-17 Stereoscopic image display device, stereoscopic image display optical element Active JP6134466B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2013054468A JP6134466B2 (en) 2013-03-17 2013-03-17 Stereoscopic image display device, stereoscopic image display optical element

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2013054468A JP6134466B2 (en) 2013-03-17 2013-03-17 Stereoscopic image display device, stereoscopic image display optical element

Publications (2)

Publication Number Publication Date
JP2014182160A JP2014182160A (en) 2014-09-29
JP6134466B2 true JP6134466B2 (en) 2017-05-24

Family

ID=51700959

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2013054468A Active JP6134466B2 (en) 2013-03-17 2013-03-17 Stereoscopic image display device, stereoscopic image display optical element

Country Status (1)

Country Link
JP (1) JP6134466B2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2853936A1 (en) * 2013-09-27 2015-04-01 Samsung Electronics Co., Ltd Display apparatus and method

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006098775A (en) * 2004-09-29 2006-04-13 Toshiba Corp Three-dimensional image display system
JP2006106607A (en) * 2004-10-08 2006-04-20 Canon Inc Image display apparatus
JP2006106608A (en) * 2004-10-08 2006-04-20 Canon Inc Image display apparatus
JP2008044136A (en) * 2006-08-11 2008-02-28 Kuraray Co Ltd Optical sheet manufacturing method and display screen manufacturing method
KR20130055997A (en) * 2011-11-21 2013-05-29 삼성디스플레이 주식회사 3-dimensional image display device

Also Published As

Publication number Publication date
JP2014182160A (en) 2014-09-29

Similar Documents

Publication Publication Date Title
JP6341903B2 (en) Image display device using diffractive lens
US20170343826A1 (en) Optical arrangement and an autostereoscopic display device incorporating the same
US8988619B2 (en) Display device with display panel forming image display and LCD panel having strip-shaped electrodes and light shielding portion
KR101392699B1 (en) Display device
JP5197852B2 (en) Stereoscopic image display device
KR101336429B1 (en) Liquid crystal display device
CN102116990B (en) Liquid crystal lens electrically driven and stereoscopic display device thereof
US10136125B2 (en) Curved multi-view image display apparatus and control method thereof
EP1394593B1 (en) 3D image / 2D image switching display apparatus
KR101269330B1 (en) Image display apparatus, display panel
US8614771B2 (en) Lenticular lens array and image display device including the same
KR100684715B1 (en) Stereoscopic image display and electronics with the same
US20130250072A1 (en) 2d/3d display and method for forming 3d image
KR101241770B1 (en) Stereo-scopic image conversion panel and stereo-scopic image display apparatus having the same
KR100598758B1 (en) 3-dimension display device
US8605227B2 (en) Image display device using diffractive device
CN1614505B (en) Three-dimensional image display device, portable terminal device and fly eye lens
US9188801B2 (en) Display apparatus and liquid crystal barrier device
US8817082B2 (en) Autostereoscopic display device
JP5465430B2 (en) Control of angle range of autostereoscopic viewing zone
US8638402B2 (en) Stereoscopic display
KR101658147B1 (en) Liquid Crystal Lens Electrically driven and Stereoscopy Display Device
JP5127530B2 (en) Stereoscopic image display device
US8279272B2 (en) Autostereoscopic display with planar pass-through
US10241342B2 (en) Stereoscopic display device and method for manufacturing the same

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20160316

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20170123

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20170131

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20170328

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20170411

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20170423

R150 Certificate of patent or registration of utility model

Ref document number: 6134466

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150