Imaging device

The imaging device achieves enhanced omnidirectional stereo imaging by arranging optical systems at polygon vertices with specific ratios and angles, addressing parallax stitching and depth perception issues in 360-degree panoramic images.

WO2026140563A1PCT designated stage Publication Date: 2026-07-02CANON KK
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
CANON KK
Filing Date
2025-11-12
Publication Date
2026-07-02

AI Technical Summary

Technical Problem

Existing imaging devices struggle to perform omnidirectional stereo imaging effectively, with issues in parallax stitching and depth perception, particularly in 360-degree panoramic images.

Method used

The imaging device employs a configuration of first and second optical systems arranged at the vertices of a polygon, with specific ratios and angles between optical axes to ensure optimal baseline length and parallax, allowing for seamless stitching and enhanced depth perception in 360-degree stereo imaging.

Benefits of technology

This configuration enables high-quality omnidirectional stereo imaging with improved stitching and depth perception, ensuring immersive and seamless panoramic images.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025039577_02072026_PF_FP_ABST
    Figure JP2025039577_02072026_PF_FP_ABST
Patent Text Reader

Abstract

[Problem] To provide an imaging device that can perform good entire-periphery stereoscopic imaging. [Solution] Provided is an imaging device wherein first pairs of optical systems, each constituted by a first optical system Rj and a second optical system Lj, are disposed so as to be respectively positioned at the vertices of a polygon having five or more sides. When IPD is defined as the distance between respective surface vertices of optical surfaces furthest to the object side of a first optical system and a second optical system which constitute a second pair for stereoscopic imaging and which are a different combination than the first pairs, and NPD is defined as the distance between respective surface vertices of optical surfaces furthest to the object side of either the first optical systems or the second optical systems of two first pairs of optical systems that are adjacent in the peripheral direction of the polygon, the expression 0.01≤NPD / IPD≤0.70 is satisfied.
Need to check novelty before this filing date? Find Prior Art

Description

Imaging device

[0001] The present invention relates to an imaging device capable of omnidirectional stereo imaging.

[0002] Patent Document 1 discloses an imaging device that performs omnidirectional stereo imaging by arranging a pair of cameras composed of a right-eye camera and a left-eye camera arranged in parallel so as to face different directions along a circle. Each pair of cameras can perform stereo imaging that reproduces the binocular parallax of a person. In such an imaging device, a plurality of images obtained by a plurality of right-eye cameras and a plurality of images obtained by a plurality of left-eye cameras are stitched together by stitching processing to generate an omnidirectional panoramic image capable of stereo viewing.

[0003] U.S. Patent No. 10,750,153

[0004] There is a need for an imaging device that can perform omnidirectional stereo imaging better than before.

[0005] An imaging device according to one aspect of the present invention is an imaging device in which a first pair of optical systems including a first optical system and a second optical system are arranged so as to be located at each vertex of a polygon having five or more sides. When the distance between the vertexes (or nodal points) of the most object-side optical surfaces of the first optical system and the second optical system forming a second pair for performing stereo imaging in a combination different from the first pair is IPD, and the distance between the vertexes of the most object-side optical surfaces of the first optical systems or the second optical systems of two adjacent first pairs of optical systems in the circumferential direction of the polygon is NPD, it is characterized by satisfying the condition of 0.01 ≤ NPD / IPD ≤ 0.70.

[0006] According to the present invention, it is possible to provide an imaging device that can obtain an imaging device capable of performing good omnidirectional stereo imaging.

[0007] Schematic diagrams of the optical systems of the imaging devices of Examples 1 to 10. Schematic diagrams of the first pair of optical systems of the imaging devices of Examples 1 to 10. Another schematic diagram of the first pair of optical systems. Diagrams showing the arrangement of the optical systems in the imaging devices of Examples 1 to 3. Diagrams showing the arrangement of the optical systems in the imaging devices of Examples 4 and 5. Diagrams showing the arrangement of the optical systems in the imaging device of Example 6. Diagrams showing the arrangement of the optical systems in the imaging device of Example 7. Diagrams showing the arrangement of the optical systems in the imaging device of Example 8. Diagrams showing the arrangement of the optical systems in the imaging device of Example 9. Diagrams showing the arrangement of the optical systems in the imaging device of Example 10. Cross-sectional view of the optical system of Numerical Example 1. Aberration diagram of the optical system of Numerical Example 1. Cross-sectional view of the optical system of Numerical Example 2. Aberration diagram of the optical system of Numerical Example 2.

[0008] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the X and Z directions are shown in the drawings, and the direction perpendicular to the X and Z directions (the direction perpendicular to the plane of the drawing) is shown as the Y direction.

[0009] Figure 1 shows the basic configuration (XZ cross-section) of the optical system 101 used in the imaging devices of Examples 1 to 10. The X direction is the direction in which the optical axis AX of the optical system 101 extends (optical axis direction), the Y direction is the horizontal direction, and the Z direction is the vertical direction. The optical system 101 has a horizontal field of view HFOV and a vertical field of view VFOV (not shown). The optical axis AX passes through the vertex of the optical surface (lens surface) closest to the object (hereinafter simply referred to as the vertex of the surface closest to the object) of the optical system 101 and is perpendicular to the image plane of the optical system 101. The imaging surface of the image sensor 102 is located on the image plane.

[0010] Each imaging device in each embodiment has five or more first pairs of optical systems, each consisting of a right-eye optical system as a first optical system and a left-eye optical system as a second optical system, each composed of the optical system 101 shown in Figure 1. These five or more first pairs of optical systems are each arranged at the vertices of a polygon with five or more sides, as will be explained later in the figures. Specifically, the vertices of the lens surfaces closest to the object for all first optical systems, the vertices of the lens surfaces closest to the object for all second optical systems, the nodal points (no-parallax points) of all first optical systems, or the nodal points of all second optical systems are arranged at the vertices of a polygon with five or more sides. An imaging device having multiple optical systems 101 arranged in this way performs 360° stereoscopic imaging around the center of the polygon.

[0011] Each optical system 101 may be referred to as an imaging device (camera), and the first pair of optical systems 101 may be referred to as the first pair of imaging devices. However, even in this case, an imaging system including multiple pairs of first imaging devices corresponds to the "imaging device" referred to in each embodiment.

[0012] In the optical system 101 shown in Figure 1, La is the distance along the optical axis from the vertex of the lens surface closest to the object to the image plane, and represents the total optical length of the optical system 101.

[0013] Figures 2 and 3 show the arrangement of the right-eye optical system 101R and the left-eye optical system 101L4 in the first pair of optical systems 103 as a first example and in the first pair of optical systems 104 as a second example, respectively. The right-eye optical system 101R and the left-eye optical system 101L are arranged such that their optical axes AX form an angle θ. The angle θ is the angle measured counterclockwise from the optical axis AX of the left-eye optical system 101L to the emitting optical axis AX of the right-eye optical system 101R. The angle θ may be 180° as shown in Figure 2 or less than 180° as shown in Figure 3, and it is preferable that it satisfies the conditions of equation (2) described later. Furthermore, let Lb be the distance between the vertices of the faces closest to the object of the first pair of right-eye optical systems 101R and left-eye optical systems 101L.

[0014] Figures 4, 5, 6, 7, 8, 9, and 10 show the arrangement of five or more first pairs of optical systems in Examples 1 to 10, respectively. In each figure, the right-eye optical system is denoted as Rj and the left-eye optical system as Lj, with the order of j counted clockwise (1, 2, 3, ...). The first pair of optical systems is a pair of right-eye optical system Rj and left-eye optical system Lj with the same j.

[0015] As mentioned above, five or more pairs of first optical systems are arranged at the vertices of a polygon with five or more sides. In other words, the lines connecting the vertices of the faces closest to the object of five or more right-eye optical systems 101Rj and the lines connecting the vertices of the faces closest to the object of five or more left-eye optical systems 101Lj are arranged to form a polygon with five or more sides. Figure 4 shows an example where five pairs of first optical systems are arranged at the vertices of a pentagon, while the other figures show N pairs of first optical systems arranged at the vertices of an N-sided polygon.

[0016] Furthermore, the imaging device in each embodiment acquires an image that has parallax between the two and allows for stereoscopic viewing (a parallax image consisting of a right-eye image and a left-eye image) by imaging with a second pair of right-eye optical systems and left-eye optical systems in a different combination than the first pair. In the following description, the second pair will also be referred to as a stereo pair. If N pairs of the first pair of optical systems are arranged, the number of stereo pairs will also be N.

[0017] The distance IPD shown in Figures 4 to 10 is the distance between the vertices of the surfaces closest to the object in the right-eye optical system and the left-eye optical system of a stereo pair, and is also called the baseline length. To enable good stereoscopic viewing of disparity images, the distance IPD is preferably set in the range of approximately 50 to 80 mm, corresponding to the approximately 65 mm distance between human eyes. However, to enhance the sense of depth for distant objects, the distance IPD may be set to 100 mm or 150 mm, etc. Thus, the distance IPD should be set according to the object being primarily imaged.

[0018] The distance NPD shown in Figures 4 to 10 is the distance between the object-side vertices of the right-eye optical systems or left-eye optical systems in two adjacent (i.e., closest) first pairs of optical systems in the circumferential direction of the polygon. To obtain a panoramic right-eye image and a panoramic left-eye image of the entire circumference, the imaging device stitches together multiple right-eye images acquired by all right-eye optical systems and multiple left-eye images acquired by all left-eye optical systems. In this case, if the distance NPD is too large, the parallax between the two images stitched together will become too large, resulting in mismatch at the seam, or requiring complex stitching to reduce this mismatch.

[0019] The distance NPD is uniquely determined by the distance (baseline length) IPD of each stereo pair, the distance Lb in each first pair of optical systems, and the arrangement of five or more first pairs of optical systems. Furthermore, it is preferable that the imaging device of each embodiment satisfies the conditions of the following equation (1).

[0020] 0.01 ≤ NPD / IPD ≤ 0.70 (1) The condition in equation (1) shows an appropriate relationship between the baseline length of the stereo pair and the distance between the two optical systems used to acquire the two images that are stitched together. Figures 4 to 10 show the IPD and NPD based on one representative first pair of optical systems (left eye optical system L1), but if N pairs of first pairs of optical systems are arranged, the IPD can be measured at N locations and the NPD can be measured at 2 × N locations. It is preferable that all IPD and NPD satisfy equation (1).

[0021] If the NPD / IPD ratio exceeds the upper limit of equation (1), the parallax between the two optical systems used to acquire the image stitched together when sufficient baseline length is ensured for the stereo pair becomes too large, leading to the aforementioned stitching problems. Furthermore, it becomes impossible to ensure sufficient baseline length for the stereo pair in order to perform good stitching, resulting in a loss of the sense of depth experienced by the viewer. In addition, for the NPD / IPD ratio to not exceed the upper limit of equation (1), the number of optical systems in the first pair must be five or more. In the case of four pairs, the NPD / IPD value must be at least 0.71 or higher.

[0022] On the other hand, if the NPD / IPD ratio falls below the lower limit of equation (1), the number of first-pair optical systems becomes very large, resulting in an enormous number of images to be stitched, making the stitching process difficult. For example, if the NPD / IPD ratio falls below 0.10, the required number of first-pair optical systems becomes 32 or more.

[0023] Furthermore, it is more preferable to set the lower limit of formula (1) to 0.05, 0.10, 0.15, 0.20, or 0.25, and the upper limit of formula (1) to 0.69 or 0.65.

[0024] Furthermore, in equation (1), the IPD and NPD are defined as the vertices of the faces closest to the object in the two optical systems, but they may be replaced with nodal points (no-parallax points). That is, the IPD may be the distance between the nodal points of the right-eye optical system and the left-eye optical system of a stereo pair, and the NPD may be the distance between the nodal points of the right-eye optical systems or the left-eye optical systems of two adjacent first pairs of optical systems in the circumferential direction of the polygon. The vertices of the faces closest to the object in the optical systems and the nodal points may coincide, or there may be a difference between them.

[0025] By satisfying the above configuration and the conditions of equation (1), it is possible to ensure sufficient baseline length for the stereo pair while performing good stitching.

[0026] In each embodiment, the angle θ between the optical axes of the first pair of right-eye optical systems and left-eye optical systems described above preferably satisfies the condition of the following equation (2).

[0027] 100° < θ < 270° (2) When θ exceeds the upper limit of equation (2), both the right-eye and left-eye optical systems of the first pair will be used to image the area inside the imaging device. The area inside the imaging device becomes an extra imaging area in 360-degree stereo imaging. Such an extra imaging area increases the horizontal field of view (HFOV) required per optical system in 360-degree stereo imaging. For example, when the number of the first pair N = 5 and θ = 270°, an HFOV of at least 200° is required. If the required HFOV is large, the number of pixels in the image acquired by the image sensor per 1° of HFOV decreases, and the angular resolution, which is an indicator of resolution, decreases. For this reason, it is preferable that θ is less than 270°.

[0028] On the other hand, if θ falls below the lower limit of equation (2), the horizontal fields of view of the first pair of right-eye and left-eye optical systems overlap. There is almost no parallax in the images taken in these overlapping horizontal fields of view, resulting in unnecessary imaging in 360-degree stereo imaging. For example, if θ = 100° when N = 5, there is an overlap of about 8° in the horizontal fields of view of the first pair of right-eye and left-eye optical systems. For this reason, it is preferable that θ be greater than 100°.

[0029] Furthermore, it is more preferable to set the lower limit of formula (2) to 110° or 120°, and the upper limit of formula (2) to 260°, 250°, or 200°.

[0030] Furthermore, in each embodiment, it is preferable that one or more first pairs of optical systems are arranged between the right-eye optical system and the left-eye optical system of the stereo pair. For example, in Figure 4, the number of first pairs of optical systems arranged between the stereo pair in the circumferential direction of the polygon is one, and Q = 1. It is preferable that Q satisfies the conditions of the following equation (3) using the Gaussian symbol [ ].

[0031] Q ≤ [N / 2] - 1 (3) In equation (3), it is preferable that Q is 1 or greater for all N. This is because when Q is 0, IPD < NPD, and for all N, the angle θ exceeds the upper limit of equation (1).

[0032] Furthermore, when Q = [N / 2] - 1, it is preferable that N is an odd number. When N is an even number when Q = [N / 2] - 1, one left-eye optical system and the other right-eye optical system of the two first pairs of optical systems located on the longest diagonal of the polygon form a stereo pair. For example, in the case of N = 6 and Q = 2 shown in Figure 6, the stereo pair is the left-eye optical system L1 and the right-eye optical system R4. In this case, when a stereoscopic image of the same imaging area is acquired with the left-eye optical system L1 and the right-eye optical system R4, the right-eye optical system R2 and the left-eye optical system L3 are reflected in the stereoscopic image. That is, if there are six right-eye optical systems and six left-eye optical systems forming the first pair in the imaging device, then a horizontal field of view of 360 / 6 = 60° is required for each optical system in order to generate a 360° horizontal panoramic right-eye image and a panoramic left-eye image. At this time, because it coincides with one side of the hexagon at one end of the horizontal field of view of the left eye optical system L1, a portion of the right eye optical system R2 is reflected in the panoramic left eye image acquired by the left eye optical system L1. Such reflections necessitate stitching that avoids the imaging area where the reflection occurs, making the stitching process more complex. For this reason, when Q = [N / 2] - 1, it is preferable that N is an odd number.

[0033] Furthermore, in each embodiment, when R is the distance between the centroid P of the polygon and the midpoint M between the vertices of the faces closest to the object, respectively, of the right-eye optical system and the left-eye optical system of the stereo pair, it is preferable that the following condition of equation (4) is satisfied.

[0034] 0.05 ≤ R・2tan(3π / 2N) / IPD < 1.00 (4) The distance R can be set arbitrarily within the range where the condition of equation (1) is satisfied.

[0035] If R・2tan(3π / 2N) / IPD exceeds the upper limit of equation (4), the number of first pairs of optical systems required to satisfy the conditions of equation (1) becomes very large, which is undesirable because it increases the number of images to be stitched. If R・2tan(3π / 2N) / IPD falls below the lower limit of equation (4), the aforementioned distance Lb becomes very small, which is undesirable because it becomes difficult to secure space for the required number of first pairs of optical systems.

[0036] Furthermore, it is more preferable to set the lower limit of formula (4) to 0.10, 0.15, or 1.16, and the upper limit of formula (4) to 0.97 or 0.95.

[0037] In each embodiment, Lc is defined as the distance between the vertices of the face closest to the object between the first and second optical systems (for example, the right eye optical system R1 and the left eye optical system L5 in Figure 4) that are closest to each other in the circumferential direction of the polygon without forming a first pair. At this time, it is desirable that the following condition of equation (5) is satisfied.

[0038] 0.10 ≤ Lb / Lc ≤ 1.00 (5) If Lb / Lc exceeds the upper limit of equation (5), the number of first pairs of optical systems required to satisfy the conditions of equation (1) becomes very large, which is undesirable because it increases the number of images to be stitched. If Lb / Lc falls below the lower limit of equation (5), the distance Lb becomes very small, which is undesirable because it becomes difficult to secure space for the required number of first pairs of optical systems.

[0039] Furthermore, it is more preferable to set the lower limit of formula (5) to 0.15, 0.20, or 0.50, and the upper limit of formula (5) to 0.95 or 0.90.

[0040] In each embodiment, it is preferable that the angle of view in the vertical direction perpendicular to the plane containing the polygon of at least one of the first pair of right-eye optical systems and left-eye optical systems (vertical field of view VFOV) is 150° or more.

[0041] This allows for the coverage of a wide imaging area in the vertical direction when the vertical field of view (VFOV) is 150° or more. In particular, when using 360° stereo images for applications such as VR (Virtual Reality), displaying an image with a wide vertical display range enables a more immersive viewing experience of stereoscopic images. If the VFOV is less than 150°, the display range of the stereoscopic image in the vertical direction becomes narrower, and the sense of immersion decreases. When stitching the bottom and top of the 360° stereo image, it is preferable that the VFOV be 180° or more, and even 190° or more. Alternatively, the VFOV of the portion excluding the bottom and top may be 180° or more, and the VFOV of the bottom and top where the members supporting the imaging device are located may be 150° or more.

[0042] Also, it is preferable that the horizontal angular field (horizontal field of view HFOV) in a direction parallel to the plane including at least one polygon of the right-eye optical system and the left-eye optical system forming the first pair is 100° or more.

[0043] Thereby, the imaging area in 360° full-circle imaging can be covered. If the HFOV is less than 100°, it becomes difficult to obtain a sufficient angular field for generating a full-circle stereo image, and there is a high possibility that the images joined by stitching processing will not be smoothly joined, which is not preferable. It is more preferable that the HFOV is 110° or more, and further preferably 120° or more.

[0044] In each embodiment, it is preferable to satisfy the conditions of the following formula (8).

[0045] 40.0 mm ≤ IPD ≤ 150.0 mm (8) The condition of formula (8) indicates the range of an appropriate baseline length of the optical systems of the stereo pair for obtaining a good stereoscopic effect from the stereo image. If the IPD exceeds the upper limit of formula (8), the parallax of a particularly close object becomes too large, making it difficult for the viewer to fuse the close object, or the close object cannot be seen with a natural stereoscopic effect, which is not preferable. If the IPD is below the lower limit of formula (8), it becomes difficult to obtain a sufficient stereoscopic effect even for a close object, which is not preferable.

[0046] It is more preferable that the lower limit of formula (8) is 45.0 mm, 50.0 mm or 60 mm, and it is more preferable that the upper limit of formula (8) is 100.0 mm, 90.0 mm or 85 mm.

[0047] In each embodiment, it is preferable to satisfy the conditions of the following formula (9).

[0048] 0.01 ≤ La / IPD ≤ 0.25 (9) The condition in equation (9) indicates an appropriate relationship between the total optical length La of the right and left eye optical systems and the baseline length. If La / IPD exceeds the upper limit of equation (9), the optical system will be visible in the 360-degree stereo imaging, which is undesirable. If La / IPD falls below the lower limit of equation (9), the total optical length of the optical system becomes very small, which is undesirable because it reduces the degree of freedom in optical design. Furthermore, if the IPD becomes very large, the parallax for near objects becomes large, making it difficult for the viewer to fuse near objects or preventing near objects from being seen with a natural sense of depth, which is also undesirable.

[0049] Furthermore, it is more preferable to set the lower limit of formula (9) to 0.03 or 0.05, and the upper limit of formula (9) to 0.23, 0.20, or 0.15.

[0050] Examples 1 to 10 will be described in detail below. Following the description of Example 10, numerical examples 1 and 2 of the optical system 101 are shown.

[0051] The imaging device of Embodiment 1 has a configuration in which five pairs of first optical systems are arranged at each vertex of a pentagon, as shown in Figure 4. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment. The vertical and horizontal fields of view of the optical system in Numerical Example 1 are 180°, and the total optical length is 7.95 mm.

[0052] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 180°, and the distance Lb is 20.00 mm.

[0053] The five stereo pairs in this embodiment are L1 and R3, L2 and R4, L3 and R5, L4 and R1, and L5 and R2. All of these stereo pairs have the same IPD and Q=1. The pair of optical systems that acquire the two images to be stitched together (third pair: hereinafter referred to as the stitch pair) are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, and R5 and R1. All of these stitch pairs have the same NPD. The IPD is 70.0 mm and the NPD is 48.5 mm.

[0054] Furthermore, the two optical systems separated by distance Lc (the fourth pair; hereafter referred to as the close-proximity pair) are L1 and R2, L2 and R3, L3 and R4, L4 and R5, and L5 and R1. In all of these close-proximity pairs, Lc is equal. Lc is 30.90 mm.

[0055] The imaging device of Embodiment 2 has a configuration in which five pairs of first optical systems are arranged at each vertex of a pentagon, as shown in Figure 4. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0056] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 140°, and the distance Lb is 18.79 mm.

[0057] The stereo pair, stitched pair, and adjacent pair in this embodiment are the same as in Embodiment 1. In this embodiment, the IPD is 70.0 mm and the NPD is 48.1 mm. The Lc is 31.65 mm.

[0058] The imaging device of Embodiment 3 has a configuration in which five pairs of first optical systems are arranged at each vertex of a pentagon, as shown in Figure 4. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0059] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 125°, and the distance Lb is 17.74 mm.

[0060] The stereo pair, stitched pair, and adjacent pair in this embodiment are the same as in Embodiment 1. In this embodiment, the IPD is 70.0 mm and the NPD is 47.8 mm. The Lc is 32.30 mm.

[0061] The imaging device of Embodiment 4 has a configuration in which six pairs of first optical systems are arranged at each vertex of a hexagon, as shown in Figure 5. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0062] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 180°, and the distance Lb is 20.00 mm.

[0063] The six stereo pairs in this embodiment are L1 and R3, L2 and R4, L3 and R5, L4 and R6, L5 and R1, and L6 and R2. All of these stereo pairs have the same IPD and Q=1. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, and R6 and R1. All of these stitch pairs have the same NPD. The IPD is 70.0 mm and the NPD is 47.3 mm.

[0064] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, and L6 and R1. In all of these adjacent pairs, Lc is equal. Lc is 28.87 mm.

[0065] The imaging device of Embodiment 5 has a configuration in which six pairs of first optical systems are arranged at each vertex of a hexagon, as shown in Figure 5. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0066] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 125°, and the distance Lb is 17.74 mm.

[0067] The stereo pair, stitched pair, and adjacent pair in this embodiment are the same as in Embodiment 4. In this embodiment, the IPD is 70.0 mm and the NPD is 46.4 mm. The Lc is 30.17 mm.

[0068] The imaging device of Embodiment 6 has a configuration in which six pairs of first optical systems are arranged at each vertex of a hexagon, as shown in Figure 6. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0069] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 150°, and the distance Lb is 19.32 mm.

[0070] In this embodiment, the stereo pairs are L1 and R4, L2 and R5, L3 and R6, L4 and R1, L5 and R2, and L6 and R3, and the IPD of all these stereo pairs is equal, and Q = 2. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, and R6 and R1. The NPD of all these stitch pairs is equal. The IPD is 70.0 mm and the NPD is 38.8 mm.

[0071] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, and L6 and R1. The Lc of all these adjacent pairs is equal. The Lc is 20.86 mm.

[0072] The imaging device of Embodiment 7 has a configuration in which eight pairs of first optical systems are arranged at each vertex of an octagon, as shown in Figure 7. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0073] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 180°, and the distance Lb is 20.00 mm.

[0074] In this embodiment, the stereo pairs are L1 and R3, L2 and R4, L3 and R5, L4 and R6, L5 and R7, L6 and R8, L7 and R1, and L8 and R2. The IPD of all these stereo pairs is equal, and Q = 1. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L7, L7 and L8, L8 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, R6 and R7, R7 and R8, and R8 and R1. The NPD of all these stitch pairs is equal. The IPD is 70.0 mm and the NPD is 46.2 mm.

[0075] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, L6 and R7, L7 and R8, and L8 and R1. The Lc of all these adjacent pairs is equal. The Lc is 27.06 mm.

[0076] As shown in Figure 8, the imaging device of Embodiment 8 has a configuration in which eight pairs of first optical systems are arranged at each vertex of an octagon. The optical systems shown in Numerical Example 2 are used for the right and left eye optical systems in this embodiment.

[0077] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 180°, and the distance Lb is 14.00 mm.

[0078] In this embodiment, the stereo pairs are L1 and R4, L2 and R5, L3 and R6, L4 and R7, L5 and R8, L6 and R1, L7 and R2, and L8 and R3. The IPD of all these stereo pairs is equal, and Q = 2. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L7, L7 and L8, L8 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, R6 and R7, R7 and R8, and R8 and R1. The NPD of all these stitch pairs is equal. The IPD is 80.0 mm and the NPD is 435.8 mm.

[0079] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, L6 and R7, L7 and R8, and L8 and R1. The Lc of all these adjacent pairs is equal. The Lc is 22.42 mm.

[0080] As shown in Figure 9, the imaging device of Embodiment 9 has a configuration in which 12 pairs of first pairs of optical systems are arranged at each vertex of a dodecagon. The optical systems shown in Numerical Example 1 are used for the right and left eye optical systems in this embodiment.

[0081] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 180°, and the distance Lb is 20.00 mm.

[0082] In this embodiment, the stereo pairs are L1 and R3, L2 and R4, L3 and R5, L4 and R6, L5 and R7, L6 and R8, L7 and R9, L8 and R10, L9 and R11, L10 and R12, L11 and R1, and L12-R3. The IPD of all these stereo pairs is equal, and Q = 1. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L7, L7 and L8, L8 and L9, L9 and L10, L10 and L11, L11 and L12, L12 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, R6 and R7, R7 and R8, R8 and R9, R9 and R10, R10 and R11, R11 and R12, and R12 and R1. The NPDs in all of these stitch pairs are equal. The IPD is 80.0 mm and the NPD is 50.6 mm.

[0083] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, L6 and R7, L7 and R8, L8 and R9, L9 and R10, L10 and R11, L11 and R12, and L12 and R1. The Lc of all these adjacent pairs is equal. The Lc is 31.06 mm.

[0084] The imaging device of Example 10 has a configuration in which 12 pairs of first optical systems are arranged at each vertex of a dodecagon, as shown in Figure 10. The optical systems shown in Numerical Example 2 are used for the right and left eye optical systems in this embodiment.

[0085] In each first pair of optical systems, the angle θ between the optical axes AX of the right and left eye systems is 125°, and the distance Lb is 10.64 mm.

[0086] In this embodiment, the stereo pairs are L1 and R3, L2 and R4, L3 and R5, L4 and R6, L5 and R7, L6 and R8, L7 and R9, L8 and R10, L9 and R11, L10 and R12, L11 and R1, and L12 and R3. The IPD of all these stereo pairs is equal, and Q = 4. The stitch pairs are L1 and L2, L2 and L3, L3 and L4, L4 and L5, L5 and L6, L6 and L7, L7 and L8, L8 and L9, L9 and L10, L10 and L11, L11 and L12, L12 and L1, R1 and R2, R2 and R3, R3 and R4, R4 and R5, R5 and R6, R6 and R7, R7 and R8, R8 and R9, R9 and R10, R10 and R11, R11 and R12, and R12 and R1. The NPDs in all of these stitch pairs are equal. The IPD is 80.0 mm and the NPD is 22.3 mm.

[0087] Furthermore, the adjacent pairs are L1 and R2, L2 and R3, L3 and R4, L4 and R5, L5 and R6, L6 and R7, L7 and R8, L8 and R9, L9 and R10, L10 and R11, L11 and R12, and L12 and R1. The Lc of all these adjacent pairs is equal. The Lc is 11.89 mm.

[0088] In Examples 1 to 10, imaging devices capable of 360° stereo imaging were shown having 5, 6, 8, and 12 pairs of the first optical system, but the number of the first optical system is not limited to these.

[0089] Examples 1 to 10 and the numerical values ​​related to the conditions described above are summarized in Table 1 below.

[0090]

[0091] Figures 11 and 13 show the optical system configurations for numerical example 1 and numerical example 2, respectively. In these figures, the left side is the object side and the right side is the image side. SP indicates the aperture diaphragm, and IP indicates the image plane. GB indicates an optical block corresponding to an optical filter, faceplate, low-pass filter, or infrared cut filter, etc.

[0092] The optical system in each numerical example consists of seven lenses. To shorten the overall optical length and reduce reflections from each lens, it is preferable that the optical system consists of seven or fewer lenses. Specifically, the optical system in each numerical example consists of two convex meniscus lenses on the object side, two positive lenses, a negative lens, a positive lens, and a negative lens, arranged in order from the object side to the image side. The use of two meniscus lenses facilitates widening the angle of view and correcting various aberrations such as field curvature. The aperture diaphragm SP is positioned between the two positive lenses, and the optical system has a symmetrical optical configuration with respect to the aperture diaphragm SP. This makes it easier to correct off-axis aberrations and ensure sufficient peripheral light.

[0093] Figures 12 and 14 show the longitudinal aberrations (spherical aberration, astigmatism, distortion, and chromatic aberration) of the optical systems for numerical example 1 and numerical example 2, respectively. In the spherical aberration diagram, Fno indicates the F number, the solid line shows the spherical aberration at the d line (wavelength 587.6 nm), and the dashed line shows the spherical aberration at the g line (wavelength 435.8 nm). In the astigmatism diagram, the solid line ΔS shows the astigmatism at the sagittal image plane, and the dashed line ΔM shows the astigmatism at the meridional image plane. The distortion diagram shows the distortion at the d line. The chromatic aberration diagram shows the lateral chromatic aberration at the g line. ω is the half-angle of view (°).

[0094] The specific numerical data for numerical examples 1 and 2 are shown below. The surface number i indicates the order of the surfaces when counted from the object side. r is the radius of curvature of the i-th surface from the object side (mm), d is the lens thickness or air gap on the optical axis between the i-th and (i+1)-th surfaces (mm), and nd is the refractive index of the optical material at the d-line between the i-th and (i+1)-th surfaces. νd is the Abbe number with respect to the d-line of the optical material between the i-th and (i+1)-th surfaces.

[0095] The Abbe number νd, with respect to the d-line, is given by νd = (Nd-1) / (NF-NC), where Nd, NF, and NC are the refractive indices of the Fraunhofer lines d-line (587.6 nm), F-line (486.1 nm), and C-line (656.3 nm), respectively.

[0096] In each numerical example, d, focal length f (mm), F-number, and half-angle of view (°) are all values ​​when the optical system is focused on an object at infinity. BF represents the back focus (mm). The back focus is the distance along the optical axis from the image-side lens surface (final surface) of the optical system to the paraxial image plane, expressed in terms of air-equivalent length. The optical block GB surface is not included as the final surface. The total lens length is the distance along the optical axis from the object-side lens surface (frontmost surface) of the optical system to the final surface plus the back focus, and corresponds to the total optical length La.

[0097] The total optical length La and the focal length f of the entire optical system preferably satisfy the following condition (10).

[0098] 2.0 ≤ La / f ≤ 5.5 (10) If La / f falls below the lower limit of equation (10), the total optical length La becomes too short, making it difficult to correct various aberrations such as spherical aberration, coma aberration, and field curvature, which is undesirable. If La / f exceeds the upper limit of equation (10), the total optical length La becomes too long, making it difficult to reduce reflections from each lens, which is also undesirable. In numerical example 1, La = 7.95 mm and f = 2.49, so La / f = 3.19. In numerical example 2, La = 4.97 mm and f = 1.55, so La / f = 3.21.

[0099] Furthermore, in the surface data for each numerical example, the "*" next to the surface number indicates that the surface has an aspherical shape. The aspherical shape is expressed by the following formula, where x is the displacement from the surface vertex in the optical axis direction, h is the height from the optical axis in the direction perpendicular to the optical axis, the direction of light propagation is positive, R is the radius of paraxial curvature, K is the cone constant, and A4, A6, and A8 are aspherical coefficients. The "e±x" of the cone constant and aspherical coefficients is ×10 ±x x = (h 2 / R) / [1+{1-(1+k)(h / R) 2} 1 / 2 ] + A4・h 4 +A6・h 6 +A8・h 8[Numerical Example 1] Unit: mm Surface Data Surface Number rd nd νd 1 16.518 0.55 1.53504 55.7 2* 1.708 1.29 3* 3.268 0.64 1.68040 18.1 4* 3.210 0.53 5* 2.009 0.58 1.54450 56.0 6* -5.213 0.05 7 (Aperture) ∞ 0.24 8* -12.630 0.50 1.54450 56.0 9* -1.581 0.04 10* -5.212 0.40 1.63277 23.3 11* 4.406 0.46 12* 4.272 0.85 1.53504 55.7 13* -8.825 0.78 14* -1.553 0.40 1.53504 55.7 15* -22.540 0.04 16 ∞ 0.20 1.51633 64.1 17 ∞ 0.40 Image plane ∞ Aspherical data Second plane K = 2.49324e-02 A 4 = 5.12098e-04 A 6 = 2.92036e-04 A 8 = 7.49998e-04 Third plane K = 3.29902e+00 A 4 = 2.73775e-02 A 6 = 5.17711e-03 A 8 = 1.36676e-03 Fourth plane K = 7.38603e+00 A 4 = 7.25329e-02 A 6 = 1.04584e-02 A 8 = 1.81091e-02 Fifth plane K = -1.28307e-01 A 4 = 2.20358e-02 A 6 = 7.40155e-04 A 8=-4.06109e-02 6th side K = 1.51271e+00 A 4=-2.55416e-02 A 6=-3.17946e-02 A 8= 4.71462e-03 8th side K = 6.24411e+01 A 4=-2.36172e-03 A 6=-5.15184e-02 9th side K = 9.25963e-01 A 4= 1.44439e-01 A 6=-2.46036e-01 A 8= 1.34628e-01 10th side K = 2.72676e+01 A 4=-5.56691e-02A 6=-1.06849e-01 A 8=-1.02783e-02 11th side K =-9.23750e+01 A 4=-3.61608e-02 A 6= 2.64295e-03 A 8= 2.42555e-03 12th side K =-3.91600e+01 A 4=-7.86181e-03 A 6= 2.89901e-03 A 8=-2.40888e-04 13th side K = 3.49187e+00 A 4=-3.43225e-03 A 6= 3.40743e-03 A 8=-2.52366e-04 14th side K =-9.87255e-01 A 4= 3.86995e-02 A 6=-7.92113e-04 A 8=-6.03662e-05 Surface 15 K = 4.27288e+01 A 4=-2.85577e-03 A 6= 5.01753e-05 A 8= 1.25099e-05 Various Data Focal Length 2.49 F-number 2.80 Half-angle of view (°) 90.0 Image height 3.89 Lens length 7.95 BF 0.40 [Numerical Example 2] Unit mm Surface Data Surface Number rd nd vd 1 10.324 0.34 1.53504 55.7 2* 1.067 0.81 3* 2.042 0.40 1.68040 18.1 4* 2.007 0.33 5* 1.255 0.36 1.54450 56.0 6* -3.258 0.03 7 (Aperture) ∞ 0.15 8* -7.894 0.31 1.54450 56.0 9* -0.988 0.03 10* -3.257 0.25 1.63277 23.3 11* 2.754 0.29 12* 2.670 0.53 1.53504 55.7 13* -5.516 0.48 14* -0.971 0.25 1.53504 55.7 15* -14.088 0.03 16 ∞ 0.12 1.51633 64.1 17 ∞ 0.25 Image plane ∞ Aspherical data Second plane K = 2.49324e-02 A 4= 2.09755e-03 A 6= 3.06221e-03 A 8= 2.01326e-02 Third plane K = 3.29902e+00 A 4= 1.12138e-01 A 6= 5.42859e-02 A 8= 3.66886e-02 Fourth plane K = 7.38603e+00 A 4= 2.97095e-01 A 6= 1.09665e-01 A 8= 4.86113e-01 Fifth plane K =-1.28307e-01 A 4= 9.02585e-02 A 6= 7.76109e-03 A 8=-1.09014e+00 6th side K = 1.51271e+00 A 4=-1.04618e-01 A 6=-3.33390e-01 A 8= 1.26557e-01 8th side K = 6.24411e+01 A 4=-9.67360e-03 A 6=-5.40210e-01 9th side K = 9.25963e-01 A 4= 5.91622e-01 A 6=-2.57988e+00 A 8= 3.61389e+00 10th side K = 2.72676e+01 A 4=-2.28021e-01 A 6=-1.12039e+00 A 8=-2.75907e-01 11th side K =-9.23750e+01 A 4=-1.48115e-01 A 6= 2.77134e-02 A 8= 6.51104e-02 12th side K =-3.91600e+01 A 4=-3.22020e-02 A 6= 3.03983e-02 A 8=-6.46628e-03 13th side K = 3.49187e+00 A 4=-1.40585e-02 A 6= 3.57295e-02 A 8=-6.77439e-03 14th side K =-9.87255e-01 A 4= 1.58513e-01 A 6=-8.30590e-03 A 8 = -1.62044e-03 15th surface K = 4.27288e+01 A 4 = -1.16972e-02 A 6 = 5.26127e-04 A 8 = 3.35809e-04 Various data: Focal length 1.55 F-number 2.80 Half-angle of view (°) 90.0 Image height 2.43 Lens length 4.97 BF 0.25

[0100] The embodiments described above are merely representative examples, and various modifications and changes can be made to each embodiment when implementing the present invention.

Claims

1. An imaging device comprising a first pair of optical systems, consisting of a first optical system and a second optical system, arranged so as to be located at each vertex of a polygon with five or more sides, wherein the imaging device satisfies the condition 0.01 ≤ NPD / IPD ≤ 0.70, where IPD is the distance between the vertices of the optical surfaces closest to the object in each of the first and second optical systems forming a second pair for stereo imaging in a combination different from the first pair, and NPD is the distance between the vertices of the optical surfaces closest to the object in each of the first optical systems or the second optical systems in two adjacent pairs of the first pair of optical systems in the circumferential direction of the polygon.

2. The imaging apparatus according to claim 1, characterized in that when θ is the angle between the optical axes of the first optical system and the second optical system in the first pair, the condition 100° < θ < 270° is satisfied.

3. The imaging apparatus according to claim 1 or 2, characterized in that one or more of the first pair of optical systems are arranged between the second pair of the first optical system and the second optical system.

4. The imaging apparatus according to claim 3, characterized in that when Q is the number of one or more first pairs of optical systems, N is the total number of first pairs of optical systems, and the Gaussian symbol is [ ], the condition Q ≤ [N / 2] - 1 is satisfied.

5. The imaging device according to claim 4, characterized in that when Q = [N / 2] - 1, N is an odd number.

6. The imaging apparatus according to any one of claims 1 to 5, characterized in that when R is the distance between the centroid of the polygon and the midpoint between the vertices of the optical surfaces closest to the object of the second pair of first and second optical systems, the condition 0.05 ≤ R・2tan(3π / 2N) / IPD < 1.00 is satisfied.

7. The imaging apparatus according to any one of claims 1 to 6, characterized in that, when Lb is the distance between the vertices of the optical surfaces closest to the object in the first optical system and the second optical system in the first pair, and Lc is the distance between the vertices of the optical surfaces closest to the object in the first optical system and the second optical system that are closest to each other in the circumferential direction of the polygon, rather than in the first pair, the condition 0.10 ≤ Lc / Lb ≤ 1.00 is satisfied.

8. The imaging apparatus according to any one of claims 1 to 7, characterized in that the angle of view in a direction perpendicular to the plane containing the polygon of at least one of the first optical system and the second optical system in the first pair is 150° or more.

9. The imaging apparatus according to any one of claims 1 to 8, characterized in that the angle of view in a direction parallel to the plane containing the polygon of at least one of the first optical system and the second optical system in the first pair is 100° or more.

10. An imaging apparatus according to any one of claims 1 to 9, characterized in that it satisfies the condition 40.0 mm ≤ IPD ≤ 150.0 mm.

11. The imaging apparatus according to any one of claims 1 to 10, characterized in that, when the total optical length of the first and second optical systems is La, the condition 0.01 ≤ La / IPD ≤ 0.25 is satisfied.

12. The imaging device according to any one of claims 1 to 11, characterized in that the vertices of the lens surfaces closest to the object in all of the first optical systems in the imaging device are located at the vertices of the polygon.

13. The imaging device according to any one of claims 1 to 11, characterized in that the vertices of the lens surfaces closest to the object of all the second optical systems in the imaging device are located at the vertices of the polygon.

14. The imaging device according to any one of claims 1 to 11, characterized in that all nodal points of the first optical system in the imaging device are located at each of the vertices of the polygon.

15. The imaging device according to any one of claims 1 to 11, characterized in that all nodal points of the second optical system in the imaging device are located at each of the vertices of the polygon.

16. An imaging device in which a first pair of optical systems, consisting of a first optical system and a second optical system, are arranged to be located at each vertex of a polygon with five or more sides, wherein the IPD is the distance between the nodal points of the first optical system and the second optical system forming a second pair for stereo imaging in a combination different from the first pair, and the NPD is the distance between the nodal points of the first optical systems or the second optical systems in two adjacent pairs of the first pair of optical systems in the circumferential direction of the polygon, and the device satisfies the condition 0.01 ≤ NPD / IPD ≤ 0.70.