Low-parallax imaging system with internal spatial frame

Abstract

A frame for a low-parallax imager includes a plurality of interconnected surfaces configured to mount cameras. The surfaces are mounted along a peripheral edge using kinematic linkages. The frame provides a structure that allows cameras mounted thereon to provide a combined field of view of up to approximately 360 degrees with minimal parallax between adjacent cameras.

Description

[Technical Field]

[0001] Cross-reference of related applications This disclosure includes (1) U.S. Provisional Patent Application No. 63 / 183,961, titled "Low Parallax Imaging System with an Internal Space Frame," filed on 4 May 2021; (2) International Patent Application No. PCT / US21 / 17284, titled "Panoramic Camera System for Enhanced Sensing," filed on 9 February 2021; (3) International Patent Application No. PCT / US20 / 66702, titled "Mounting Systems for Multi-Camera Imagers," filed on 22 December 2020; (4) International Patent Application No. PCT / US20 / 39197, titled "Opto-Mechanics of Panoramic Capture Devices with Abutting Cameras," filed on 23 June 2020; and (5) "Multi-camera Panoramic Image Capture Devices with a Faceted (6) The applicant claims priority to International Patent Application PCT / US20 / 39200, titled “Dome”, and to International Patent Application PCT / US20 / 39201, titled “Lens Design for Low Parallax Panoramic Camera Systems”, filed on 23 June 2020. The last four listed international applications each claim priority to U.S. Provisional Patent Application 62 / 952,973, titled “Opto-Mechanics of Panoramic Capture Devices with Abutting Cameras”, filed on 23 December 2019, and to U.S. Provisional Patent Application 62 / 952,983, titled “Multi-camera Panoramic Image Capture Devices with a Faceted Dome”, filed on 23 December 2019. Each of the last three international applications listed above also claims priority to U.S. Provisional Patent Application No. 62 / 865,741, filed on June 24, 2019.The entirety of each of the international applications listed above and each of the U.S. provisional patent applications is hereby incorporated by reference into this specification.

[0002] The present disclosure relates to a panoramic low-parallax multi-camera capture device having a plurality of adjacent polygonal cameras. The present disclosure also relates to an opto-mechanical design of a camera that captures incident light from a polygonal field of view to form a polygonal image, particularly versions thereof that use an internal space frame.

Background Art

[0003] Panoramic cameras have substantial value due to their ability to simultaneously capture wide-angle images. The oldest such example is the fisheye lens, which is an ultra-wide-angle lens that produces strong visual distortion while capturing a wide panoramic or hemispherical image. The field of view (FOV) of a fisheye lens is typically between 100 and 180 degrees, but this approach has been extended to even wider angles, including ranges up to 220 - 270°, as provided by Y. Shimizu in U.S. Patent 3,524,697. Alternatively, there are mirror or reflective-based cameras that capture annular panoramic images, such as the system proposed by P. Greguss in U.S. Patent 4,930,864. Although these technologies continue to evolve, it is difficult for them to provide complete hemispherical or spherical images with the resolution and image quality that current applications are seeking.

[0004] As an alternative, panoramic multicamera systems, featuring multiple cameras positioned around or around a sphere, are becoming increasingly common. However, in most of these systems, including those described in U.S. 9,451,162 and U.S. 9,911,454 to A. Van Hoff et al. of Jaunt Inc., the multiple cameras are sparsely located on the outer surface of the device. To capture a complete 360-degree panoramic image, including gaps or seams between adjacent individual cameras, the cameras consequently have an expanded FOV that overlaps each other. In some cases, as much as 50% of the camera's FOV or resolution may be used for camera overlap, which also creates a substantial parallax difference between the captured images. Parallax is the visual perception that the position or orientation of an object appears different when viewed from different positions. Subsequently, in image processing, both the extra image overlap and parallax difference complicate and significantly delay the effort to properly combine, tile, or stitch together an acceptable image from the images captured by adjacent cameras.

[0005] There are also panoramic multi-camera systems in which multiple cameras are arranged around a sphere or its circumference such that adjacent cameras touch along part or all of the adjacent edges. For example, U.S. Patent 7,515,177 by K. Yoshikawa describes an imaging device with a number of adjacent image pickup units (cameras). Images are collected from cameras with overlapping fields of view to correct for mechanical errors.

[0006] More broadly, in multi-camera systems, mechanical variations in assembly and alignment of individual cameras and adjacent cameras relative to each other can cause actual physical variations in the cameras themselves, as well as in the seam width and the parallelism of the camera edges along the seams. These variations can then affect the field of view (FOV) captured by individual cameras, the parallax error in the images captured by adjacent cameras, the extent of the "blind spot" in the FOV corresponding to the seam, the seam width, and the amount of image overlap that needs to be corrected. Thus, there are opportunities to improve panoramic multi-camera systems and their low-parallax cameras, as well as in their optical and optomechanical designs, and in other aspects. [Brief explanation of the drawing]

[0007] [Figure 1] This shows a part of the multi-camera system's acquisition device, and specifically, a 3D view of its two adjacent cameras. [Figure 2A] A cross-sectional view of a camera lens assembly, including the lens element and light ray path, is shown. [Figure 2B] A cross-sectional view of a camera lens assembly, including the lens element and light ray path, is shown. [Figure 3] This shows a cross-sectional view of a typical multi-camera acquisition system, illustrating the FOV overlap, field of view, overlap, seam, and blind areas. [Figure 4] The diagram shows two polyhedron shapes, a regular dodecahedron and a truncated icosahedron, for which multi-camera acquisition devices can be designed and manufactured. [Figure 5A] This shows the optical geometry with respect to the field of view for adjacent hexagonal and pentagonal lenses, which can occur in a device having a truncated icosahedron geometry. Figure 5B shows a more detailed enlarged area of ​​Figure 5A. [Figure 5B] This shows the optical geometry with respect to the field of view for adjacent hexagonal and pentagonal lenses, which can occur in a device having a truncated icosahedron geometry. Figure 5B shows a more detailed enlarged area of ​​Figure 5A. [Figure 5C]Figure 5C shows an example of low parallax (LP) quantities located near the paraxial NP point or both the entrance pupil and the device center. [Figure 5D] The parallax difference between two adjacent cameras is shown relative to the center of perspective. [Figure 5E] This shows the front color at the edge of the outer compressor lens element. [Figure 6] The graph shows the distortion correction curve plotted on a graph representing the percentage of distortion relative to the fractional field of view. [Figure 7] The field of view for adjacent cameras, including both the core and extended field of view (FOV), can be useful for designing an optimized panoramic multi-camera acquisition system. [Figure 8] This demonstrates an improved design for low parallax camera lenses or objective lenses equipped with multi-compressor lenses. [Figure 9] This paper demonstrates an improved camera lens design that functions as an objective lens, combined with a refractive relay optical imaging system. [Figure 10] This shows an electronic system diagram for a multi-camera acquisition device. [Figure 11] This illustrates the concept of an internal spatial frame that can be used to mount camera channels to a multi-camera acquisition system. [Figure 12A] This section presents the concept of a kinematic spatial frame. [Figure 12B] Side and perspective views of kinematic elements that can be used in a kinematic spatial frame are shown. [Figure 12C] Side and perspective views of kinematic elements that can be used in a kinematic spatial frame are shown. [Figure 12D] Perspective views of kinematic elements usable in a kinematic spatial frame are shown. [Figure 12E] Two perspective views of a partial assembly of the kinematic spatial frame are shown. [Figure 12F] A third perspective view of a partial assembly of the kinematic spatial frame is shown. [Figure 12G]Shows another perspective view of a partial assembly of a kinematic space frame. [Figure 13] Figs. 13A - C. Each shows an alternative kinematic space frame structure. [Figure 14] Figs. 14A - C. Show aspects of the assembly of the space frame of Fig. 14A. [Figure 15A] Shows both a perspective view and a side view of a portion of a camera lens housing and features for joining it to a space frame facet. [Figure 15B] Shows the assembly of a camera lens housing to a space frame. [Figure 15C] Shows the assembly of a camera lens housing to a space frame. [Figure 15D] Shows a second example of the assembly of a camera channel lens housing to a space frame.

DETAILED DESCRIPTION OF THE INVENTION

[0008] As generally understood in the field of optical systems, a lens or lens assembly typically includes a system or device having a plurality of lens elements that are mounted within a lens barrel or housing and work together to generate an optical image. An imaging lens captures a portion of the light coming from one or more objects that exist within an object space at a distance (s) from the lens system. The imaging lens can then form an image of these objects on an output “plane,” and the image has a finite size determined by the magnification as determined by the focal length of the imaging lens and the conjugate distances to the object(s) and the imaging plane with respect to that focal length. The amount of image light passing through the lens from the object to the image is mostly determined by the size of the aperture stop of the imaging lens, which is typically quantified by one or more values for the numerical aperture (NA) or F - number (F♯ or F / ♯).

[0009] The image quality provided by an imaging lens is determined by a number of characteristics of the lens design, including the selection of the optical materials used in the design, the size, shape (or curvature) and thickness of the lens elements, the relative spacing between the lens elements, the spectral bandwidth, polarization, the optical loading (power or flux) of the transmitted light, light diffraction or scattering, and / or lens manufacturing tolerances or errors. Image quality is typically described or quantified in terms of lens aberrations (e.g., spherical, coma, or distortion), or the relative size of the resolution points provided by the lens, which is often also quantified by the modulation transfer function (MTF).

[0010] In a typical electronic or digital camera, an image sensor is nominally placed at the image plane. This image sensor is typically a CCD or CMOS device, which is physically attached to a heat sink or other heat removal means, and also includes electronics for powering the sensor, and readout and communication circuitry for providing the image data to data storage or image processing electronics. The image sensor typically has a color filter array (CFA), such as a Bayer filter within the device, which positions color filter pixels in alignment with the image pixels to provide an array of RGB (red, green, blue) pixels. Alternative filter array patterns, including CYGM filters (cyan, yellow, green, magenta) or RGBW filter arrays (W = white), can be used instead.

[0011] In typical use, many digital cameras are used relatively independently by people or remote systems to capture images or photographs of a scene, without any dependency or interaction with any other camera device. In some cases, such as surveillance or security, camera operation may be directed by people or algorithms based on image content seen from another camera that has already captured overlapping, adjacent, or nearby image content. In another example, people capture a panoramic image of a scene, such as a landscape scene, with an expanded or wide-angle FOV by capturing a series of adjacent images in sequence, while manually or automatically moving or panning to assemble adjacent images. Image processing software such as Photoshop or Lightroom can then be used to stitch the adjacent images together, mosaic, or tile them to represent a larger expanded scene. Image stitching or photo stitching is the process of combining multiple photographic images with overlapping fields of view to produce a segmented panorama or high-resolution image. Image quality improvements, including exposure or color correction, can also be applied in real time, or in the post-processing or image rendering phase, or a combination thereof.

[0012] Unless objects in a scene are not illuminated directionally and / or have a directional light response (e.g., reflection), the available light is plenoptic, meaning that within a given space or environment, there is light that travels in all directions, or nearly all directions. The camera can then sample a subset of this light as image light, which it uses to provide a generated image showing a given view or perspective of different objects in the scene at one or more points in time. If the camera is moved to a different nearby position to capture another image of the same part of the scene, both the apparent perspective and relative position of the objects change. In the latter case, one object may now partially obscure another, while an object that was previously hidden may now be at least partially visible. These differences in the apparent position or orientation of objects are known as parallax. Specifically, parallax is the displacement or difference in the apparent position of an object viewed along two different lines of sight, measured by the angle or half-angle of the inclination between these two lines.

[0013] In stereoscopic image capture or projection systems, dual-view parallax, along with shading, occlusion, and perspective, serves as a cue to provide a sense of depth. For example, in a stereoscopic (3D) projection system, polarized or spectrally encoded pairs of images are projected onto a screen and can be seen by a viewer wearing appropriate glasses. The amount of parallax can have an optimal range; outside of this range, the resulting sense of depth is either too small to be noticed by the viewer or too large to be properly fused by the human visual system.

[0014] On the other hand, in panoramic image capture applications, parallax difference can be considered an error that can complicate both the stitching and the apparent appearance of the images. In examples where a series of panoramic landscape images are captured individually and manually, if objects in the scene are sufficiently far away (e.g., optically at infinity), the visual difference in perspective or parallax across the images may be too small to notice. Panoramic capture devices integrated with rotating cameras or multi-cameras have the potential to capture real-time image data continuously at high resolution without relying on the uncertainties of manual capture. However, such devices may also incorporate their own binocular parallax, image artifacts, or errors, including those of parallax, perspective, and exposure. While the resulting images can often be successfully stitched together using image processing algorithms, input image errors complicate image processing, increase processing time, and at the same time, sometimes leave visually apparent residual errors.

[0015] To provide context, Figure 1 shows an improved integrated panoramic multi-camera capture device 100, which has two adjacent cameras 120 within a housing 130, designed for reduced parallax image capture. These cameras are substituted for image pickup units, or camera channels, or objective lens systems. Each camera 120 has multiple lens elements (see Figure 2) mounted within a lens barrel or housing 130. The adjacent outer lens elements 137 have adjacent tapered edges 132, and one camera channel is positioned close to another camera channel, but it may not be in contact and is therefore separated by a finite-width gap or seam 160. A portion of the available light (λ), or rays 110, from the scene or object space 105 enters the camera 120 and becomes image light that is captured within a constrained FOV and directed towards the imaging plane, while other rays do not reach the camera at all. Some of the light rays 110 propagate within the camera and pass through the lens elements, forming the principal ray 170 at the edge of the field of view or peripheral rays, while other rays may potentially propagate through the lens elements, producing stray light or ghosting and false bright spots or images. For example, some of the light rays (167) that enter the outer surface of the outer lens element 137 at a large angle can travel through the camera's lens elements along complex paths, potentially producing ghost images detectable at the imaging plane 150.

[0016] More specifically, Figure 2A shows a cross-section of a portion of camera 120 having a set of lens elements 135 mounted within a housing (130, not shown) within a portion of the integrated panoramic multi-camera capture device 100. A fan of light rays 110 from object space 105 spreads across the range of principal rays from on-axis to off-axis full field of view, enters the outer lens element 137, is refracted and transmitted inward. Refractionated, passing through a further inner lens element 140 and transmitted through an aperture diaphragm 145, this image light 115 converges into a focal image at or near the image plane 150, where an image sensor (not shown) is typically positioned. The lens system 120 in Figure 2A can also be defined as having a lens configuration consisting of an outer lens element 137 or compressor lens element and an inner lens element 140, the latter of which can also be defined as consisting of a pre-stop wide-angle lens group and a post-stop eyepiece-like lens group. This compressor lens element (137) sharply directs the image light 115 inward, compressing the light and allowing the entire lens assembly to provide a short focal length. At the same time, it helps to provide the necessary space for the camera lens housing or barrel to provide the mechanical features required both to hold or mount the lens element and to interact appropriately with the adjacent camera barrel or housing. The image light that passes through the camera lens assembly from the outer lens element 137 to the image plane 150 provides an image with image quality that can be quantified by image resolution, image contrast, depth of field, and other attributes, which are defined by optical aberrations (e.g., astigmatism, distortion, or sphere) and chromatic or spectral aberrations encountered by the transmitted light in each of the lens elements (137, 140) in the camera 120. Figure 2B shows the fan of the principal ray 170, or peripheral ray, incident along or near the sloping edge 132 of the outer lens element 137 of the camera optical system (120) shown in Figure 2A. Figure 2B also shows a portion of the captured, polygonal or asymmetrical FOV 125, extending from the optical axis 185 to a line coinciding with the edge rays.

[0017] In the camera lens design shown in Figure 2A, the outer lens element 137 functions as a compressor lens element that redirects the transmitted image light 115 toward the second lens element 142, which is the first lens element in a group of inner lens elements 140. In this design, the second lens element 142 has a concave shape, very similar to the outer lens element used in fisheye-type imaging lenses. This compressor lens element helps to sharply direct the image light 115 inward or bend the light rays, allowing the entire lens assembly to provide a short focal length, while at the same time providing the necessary space for the camera lens housing 130 or barrel to provide the mechanical features necessary both to hold or mount the lens element 135 and to interact properly with adjacent camera barrels or housings. However, with good lens and optomechanical design, as well as appropriate sensor selection, the camera 120 can be designed with lens assemblies that support image resolutions of 20-30 pixels / degree, 110 pixels / degree or more, depending on the application and device configuration.

[0018] The resulting image quality from these cameras is also determined by the light scattered on the surface or within the lens elements, and the light reflected or transmitted at each lens surface. Surface transmittance and camera lens system efficiency can be improved by the use of anti-reflective (AR) coatings. Image quality can also be determined by the results of non-image light. Revisiting Figure 1, the rest of the available light can be reflected primarily by the outer lens element 137. Further light entering camera 120 can be blocked or absorbed by the aperture diaphragm or a blackened region (not shown) provided near it, the inner lens barrel surface, the edge of the lens elements, the inner baffle or light trapping features, the field diaphragm, or any combination of other surfaces. Further light entering the camera can also be stray light or ghost light potentially visible at the image plane.

[0019] The overall image quality obtained by multiple adjacent cameras 120 in an improved integrated panoramic multi-camera capture system 100 (e.g., Figure 1) may also be determined by a variety of other factors, including variations between cameras in focal length and / or track length, as well as the magnification provided by each individual camera. These parameters may vary depending on factors including variations in glass refractive index, variations in lens element thickness and curvature, and variations in lens element mounting. As an example, images tiled or mosaiced together from multiple adjacent cameras typically need to be corrected one after the other to compensate for image size variations due to camera magnification differences (e.g., ±2%).

[0020] Images generated by multiple cameras within the integrated panoramic multi-camera capture device 100 can vary in image quality and other aspects resulting in image mosaicing or tiling. Specifically, the directional pointing or collection of image light passing through the lens element to the image sensor of any given camera 120 may result in the camera being angularly distorted or asymmetrical.

number

[0021] In comparison to the system in Figure 1, a typical commercially available panoramic camera has a complete gap between cameras, where the seam can be 30-50 mm wide or more. Specifically, as shown in Figure 3, the panoramic multi-camera capture device 101 may have adjacent cameras 120 or camera channels separated by a large gap or seam 160, between which there is a blind spot or blind area 165 where no camera can capture an image. The actual physical seam 160 between adjacent camera channels or outer lens elements 137 (Figures 1 and 3) can be measured in various ways, such as the actual physical distance between adjacent lens elements or lens housings, the angular range of the lost FOV, or the number of "lost" pixels. However, the optical seam can be even larger, as the distance between the outer principal rays of one camera and another, due to any gaps in light reception caused by vignetting or coating limitations. For example, anti-reflective (AR) coatings are typically not applied to the edges of the optical system, and an offset margin is provided to provide a coated aperture (CA).

[0022] To compensate for both camera misalignment and large seams 160, and to reduce the size of blind areas 165, a typical panoramic multi-camera capture system 101 (Figure 3) has each of the individual cameras 120 capture image light 115 from a wide FOV 125 that provides overlap 127, thereby reducing the blind areas 165 and minimizing the loss of potentially captureable image content. As another example, in most commercially available multi-camera capture systems 101, the gaps are 25-50+ mm wide, and the corrected FOV overlap between cameras is similarly large, for example, the portion of FOV 125 that overlaps and is captured by two adjacent cameras 120 may be 10-50% of the camera's FOV. The presence of such large image overlap from a shared FOV 125 wastes potential image resolution, increases image processing and image stitching time, and simultaneously introduces significant image parallax and perspective errors. These errors complicate image stitching because they need to be corrected or averaged during the stitching process. In such systems, parallax is unpredictable because it changes as a function of object distance. If object distance is known, parallax can be predicted for a given field of view and camera spacing. However, since object distance is typically unknown, parallax errors consequently complicate image stitching. Optical flow and general stitching algorithms determine object depth to enable image stitching, but this comes with processing and time burdens.

[0023] Similarly, in a panoramic multi-camera capture device 100 of the type shown in Figure 1, with closely integrated cameras, the width and structure of the seam 160 can be a critical factor in the overall operation of the device. However, the seam can be made smaller than that in Figure 3, using the effective optical seam between the FOV edges of two adjacent cameras, determined by both optical and mechanical contributions. For example, by using standard optical engineering techniques to construct the lens assembly within the housing, the mechanical width of the seam 160 between the outer lens elements 137 of adjacent cameras can be reduced to 4-6 mm. For example, it is a standard technique to mount the lens elements, specifically near the outermost lens elements, within a lens barrel or housing with a minimum radial width of 1-1.5 mm. Then, consider standard coated apertures or coating margins, and possible vignetting, entrance pupil aberrations, front color, tip edge, and attempt to mount adjacent lens assemblies or housings in close proximity using standard techniques. Therefore, when considering both the optical system and the mechanism, the optical seam width between adjacent lenses can easily be 8-12 mm or more.

[0024] However, an improved version of the panoramic multi-camera capture device (300) of the type shown in Figure 1 is possible, with optical and optomechanical designs that allow for much smaller seams and further improved parallax performance. As a first example, with respect to the present technique for an improved polygonal camera, during the initial stages of manufacturing the outer lens elements 137, these lenses may have a circular shape and be AR coated up to or near their physical edges. If these lenses are subsequently machined to add a polygonal shape defining a tapered edge 132 (e.g., Figure 2B), the AR coating may essentially extend to the tapered lens edge. An effective optical or coated aperture can then be defined by mechanical mounting or any tolerance to standard edge polishing used in optical system manufacturing to avoid edge chipping. With this approach, and in combination with other techniques described later, optical seams can be reduced to a width of 1-5 mm.

[0025] Aspects of this disclosure involve generating high-quality, low-parallax panoramic images from an improved multi-camera panoramic capture system (300), a first example of which is shown in Figures 8 and 9. This broad objective can be achieved by developing a systematic range of design strategies, as well as strategies for improved image capture and processing, to inform both optical and optomechanical lens design efforts, as well as optomechanical device design and manufacturing efforts. This objective can also be achieved by providing both initial and ongoing camera and device calibrations. Broadly speaking, image processing or image rendering is a method for generating quality images from raw captured image data determined by camera intrinsics (geometric factors such as focal length and distortion), camera extrinsics (geometric factors such as camera orientation relative to object space), other camera parameters such as vignetting and transmittance, and illumination parameters such as color and directivity. With respect to the improved multi-camera panoramic capture device 300, the use of criteria in determining and tracking the central pixel or image centroid, exposure compensation, and knowledge of the camera intrinsic properties for any given camera 320 within the device all contribute to the reliable and repeatable completion of tiling of images acquired from multiple adjacent cameras. Accordingly, the following description broadly focuses on providing optical (camera or objective lens) designs that can enable the desired image quality, as well as camera and device assembly approaches, key tolerance management, camera calibration, knowledge of camera intrinsic and extrinsic properties, and other factors that may similarly affect the resulting device performance. The improved panoramic multi-camera capture device of the present invention can be used to support a wide variety of applications or markets, including cinematic image capture, augmented reality or virtual reality (VR) image capture, surveillance or security imaging, sports or event imaging, mapping or photogrammetry, vehicle navigation, and robotics.

[0026] Before investigating optomechanical means to enable an improved panoramic multi-camera acquisition device (300), means are developed to provide an improved camera 120 for use in these systems. Accordingly, the goal is to provide an improved camera (320) having both reduced parallax error and image overlap. In one aspect of this approach, the goal is to reduce residual parallax error for edge principal rays individually collected by each camera in an adjacent pair. Parallax error is defined as the variation in parallax with respect to object distance (e.g., the principal ray trajectory is slightly different with respect to a closer distance from the device (e.g., 3 feet) compared to a farther distance (e.g., 1 mile)). For example, one goal or target for reduced parallax, or effectively eliminating parallax error, or making it "parallax-free," is that the principal rays of adjacent cameras should deviate from parallel by only ≤0.5 to 2.0 degrees, and preferably ≤0.01 to 0.1 degrees. Alternatively, or equivalently, since parallax error is evaluated as a perspective error with respect to position on the image plane, it should be reduced to ≤2 pixels, and preferably ≤0.5 pixels. In another aspect of this approach, the width of the seam 160 between adjacent cameras (e.g., 120, 320) assembled within their own lens housings should be reduced. The goal is to reduce the width of the seam with respect to both their absolute physical width and their optical or effective width. For example, the goal is to reduce the seam 160 between adjacent outer lens elements 137 to have a maximum gap or actual physical seam in the range of only about 0.5 to 3.0 mm, and then reduce the maximum optical seam width to the range of only about 1 to 6 mm. As an example, these reduced seam widths can be translated into a reduced angular range of lost FOV of only 0.25 to 1.0°, or a “lost” number of pixels of only 2 to 20 pixels. For example, in the case of a device that provides 8k pixels around an equirectangular image such as a 360-degree panorama, residual image artifacts may be difficult to notice, so a loss of only 2 to 4 pixels at the seam may be acceptable.The actual details or numerical targets for virtually no parallax error or maximum optical seam width depend on many factors, including the detailed optomechanical design of the improved camera 320 and the entire apparatus 300, tolerance control, possible tolerances for the amount of center offset distance or extended FOV (215) and targets for low parallax, as well as the specifications of the entire apparatus (e.g., diameter, sensor resolution or imaging FOV or sensor pixels used within the core FOV 205 (Figure 7)). A further goal, made possible by any combination of the aforementioned improvements, is that each camera reliably and quickly provides output images from an embedded sensor package, which are cropped to provide a core FOV image, and each cropped image can then be easily stitched together or tiled with cropped images provided by adjacent cameras, so that a panoramic output image can be easily provided in real time from the improved multi-camera capture apparatus (300).

[0027] In this application to an improved multi-camera capture device (300), the improved camera 320 provided herein includes a camera lens or lens system consisting of multiple lens elements for providing images, and a camera lens housing for supporting the lens elements and junctioning them with a support structure (e.g., a spatial frame). The camera (320) is also referred to as a camera lens (320) or camera channel (320).

[0028] An improved panoramic multi-camera capture device 300, such as that shown in Figure 15C, may have multiple camera channels 320 arranged around a sphere to capture a 360-degree annular FOV, including the one proposed in Figure 15C with camera channels 920 or lens housings 905. Alternatively, a panoramic multi-camera capture device may have multiple cameras arranged around a spherical or polyhedral shape. A polyhedron is a three-dimensional solid consisting of a set of polygons adjacent at their edges. One polyhedral shape, as shown in Figure 4, is that of a dodecahedron 50, having 12 sides or faces and 20 vertices or angles (e.g., vertices 60), each formed as a regular pentagon 55. A panoramic multi-camera capture device formed in a dodecahedral shape has a camera with a pentagonal outer lens element that images at a nominal full wide-angle of 69.1°. Another shape is that of a truncated icosahedron, resembling a soccer ball, which, as shown in Figure 4, has a combination of 12 regular pentagonal sides or faces, 20 regular hexagonal sides or faces, 60 vertices, and 90 edges. More complex shapes with more faces, such as regular polyhedra, Goldberg polyhedra, or shapes with octagonal sides, or even some irregular polyhedron shapes may be useful. For example, the Goldberg chamfered dodecahedron is similar to a truncated icosahedron and has a total of 42 sides with both pentagonal and hexagonal facets. However, generally, the preferred polyhedra for this purpose have hexagonal or pentagonal sides or faces, which are generally rounded in shape and have 132 obtuse-angled tapered edges. Other polyhedron shapes, such as octahedrons or regular icosahedrons, which have triangular facets, can also be used. Polyhedral facets with steeper or sharper angles, such as square or triangular facets, may be easier to manufacture than pentagonal and / or hexagonal facets because they require less cutting to provide the polygonal edges on the outermost lens element to define the captured polygonal FOV. However, due to their sharp angles, extra care may be required when cutting, chamfering, and machining the optical system.Additionally, designing camera lenses and lens housings for optical and optomechanical performance can be even more challenging with respect to lens facets that have large FOVs and sharp facet angles. Typically, a 360° polyhedron camera does not capture a perfect spherical FOV, as at least a portion of one facet is sacrificed to allow for support features, such as mounting struts, as well as power and communication cable routing. However, if the device communicates wirelessly and is also suspended at the vertices by thin cables, the FOV lost due to such physical connections can be reduced.

[0029] As shown in Figures 1 and 2B, the camera channel 120 resembles a frustum of a cone, or a part thereof, which is a geometric solid (usually a cone or pyramid) that exists between one or two parallel faces that intersect it. In this context, the fan of the principal ray 170 corresponding to the edge of the polygon is refracted by the outer compressor lens element 137 and nominally coincides with the edge of the frustum of the polyhedral geometry.

[0030] To aid in illustrating several issues related to camera geometry, Figure 5A shows cross-sections of a pentagonal lens 175 capturing a pentagonal FOV 177 and a hexagonal lens 180 capturing a hexagonal FOV 182, representing a pair of adjacent cameras whose outer lens elements have a pentagonal and hexagonal shape, as can occur in a truncated icosahedron, or soccer ball-shaped panoramic multi-camera capture device (e.g., 100, 300). The logical hexagonal FOV 182 extends to a half-FOV of 20.9° or a full FOV of 41.8° (θ1) along the sides, although the FOV is even larger near the vertices. The pentagonal FOV 177 supports a 36.55° FOV (θ2) within a circular region, with the FOV being even larger near the corners or vertices. In particular, in this cross-section, the pentagonal FOV 177 is asymmetrical, supporting a 20-degree FOV on one side of the optical axis 185 and only a 16.5-degree FOV on the other side of the optical axis.

[0031] Optical lenses are typically designed using programs such as ZEMAX or Code V. The success of the design typically depends in part on the selection of the best or most appropriate lens parameters, which are identified as operands for use in the merit function. This is also true when designing a lens system for an improved low-parallax multi-camera panoramic capture device, for which there are several factors affecting performance (particularly parallax) and several parameters that can be optimized individually or as a whole to control it. One approach is to aim for the optimization of the "NP" point, or more meaningfully, a variation thereof.

[0032] As background, in the field of optics, there is the concept of the entrance pupil, which is the projected image of the aperture diaphragm as seen from object space, or a virtual aperture from which imaging rays from object space appear to propagate toward it before any refraction by the first lens element. By standard technique, the position of the entrance pupil can be found by identifying the paraxial principal ray from object space 105 that passes through the center of the aperture diaphragm, and projecting or extending that object space toward the position where it strikes the optical axis 185. In optics, the incident Gaussian or paraxial ray is understood to be within an angular range of 10° or less from the optical axis, corresponding to the ray directed toward the center of the aperture diaphragm, and this also defines the position of the entrance pupil. Depending on the lens characteristics, the entrance pupil may be larger or smaller than the aperture diaphragm and may be located in front of or behind the aperture diaphragm.

[0033] In contrast, in the field of low parallax cameras, there is the concept of a no-parallax (NP) point, or viewpoint center. Conceptually, an "NP point" is associated with a high-FOV chief ray or principal ray incident on or near the outer edge of the outermost lens element, projecting or extending its object space toward the position where it strikes the optical axis 185. For example, depending on the design, camera channels in a panoramic multi-camera capture device can support a half-FOV with a non-paraxial principal ray at an angle of >31° for dodecahedral systems (Figure 4) or >20° for truncated icosahedral systems (see Figures 4 and 5A). This concept of NP point projection is applied to the design of panoramic multi-camera capture devices in relation to expectations of principal ray propagation and parallax control for adjacent optics (cameras). If a camera appears to orbit around the NP point, or if multiple cameras appear to rotate around a common NP point, parallax errors are reduced, and it can be said that images can be aligned with little to no parallax or perspective error. However, in the field of low parallax cameras, the NP point is also identified with the entrance pupil, and the axial orientation of the entrance pupil is estimated using the tangent relationship of the primary optical system between the projection of the paraxial field of view in the first lens element (see Figures 2A and 2B) and the height of the incident ray.

[0034] Therefore, confusingly, in the field of low parallax camera design, NP points were previously associated with both the projection of the edge of the FOV principal ray and the projection of the principal ray within the Gaussian or paraxial region. As you can see, in practice, both have value. Specifically, NP points associated with the paraxial entrance pupil can be useful when developing initial specifications for designing and describing lenses. NP points associated with the non-paraxial edge field principal ray can be useful when targeting and understanding parallax performance and when defining the conical volume or frustum within which a lens assembly can exist.

[0035] The projection of these non-paraxial principal rays can miss the entrance pupil defined by the paraxial principal rays due to both lens aberrations and practical geometric factors associated with these lens systems. Related to the former, in well-designed lenses, image quality at the image plane is typically prioritized by limiting the effects of aberrations on resolution, telecentricity, or other attributes. Within a lens system, aberrations at the interim surfaces, including the aperture diaphragm, can vary widely, as the net sum at the image plane is the primary consideration. While aberrations at the aperture diaphragm are often somewhat controlled to avoid vignetting, non-paraxial principal rays do not need to pass through the aperture diaphragm or the center of the entrance pupil positioned in the projection paraxial direction.

[0036] To further elaborate on these concepts and to enable the design of improved low-parallax lens systems, note that the camera lens system 120 in Figure 2A shows both a first NP point 190A corresponding to the entrance pupil defined by the vector projection of a paraxial principal ray from object space 105, and an offset second NP point 190B corresponding to the vector projection of a non-paraxial principal ray from object space. Both of these ray projections intersect the optical axis 185 at a position behind both the lens system and the image plane 150. As will be described later, the ray behavior between and within the region adjacent to the projection points 190A and 190B can be complex, and neither the projection position nor the projection point has a definitive value and size. While the projection of a principal ray intersects the optical axis at a certain point, the projection of a group of principal rays converges toward the optical axis and intersects at a different position, which can be tightly grouped together (e.g., within a few microns or tens of microns), in which case the range or size of that “point” can be determined by the set of adjacent principal rays used in the analysis. On the other hand, when designing a low parallax imaging lens to image a large FOV, the axial distance or difference between NP points 190A and 190B provided by the projected paraxial and non-paraxial principal rays can be significantly large (e.g., millimeters). Thus, as will be discussed later, the axial difference represents a useful means of parallax optimization (e.g., low parallax amount 188) of the lens system designed for current panoramic capture devices and applications. As will be seen as well, the design of an improved device (300) can be optimized to position the geometric center of the device, or the device center 196, outside but close to or alternatively within this low parallax amount 188, and preferably close to the non-paraxial principal ray NP point.

[0037] In one embodiment, Figure 5A shows the projection of the theoretical edge of the field of view (FOV edge 155) past the outer lens elements (lenses 175 and 180) of two adjacent cameras to provide a line pointing to a common point (190). These lines represent the theoretical limits of complex “conical” optomechanical lens assemblies, which are typically pentagonal or hexagonal cones that limit the quantity. Again, ideally, in a parallax-free multi-camera system, the entrance pupils or NP points of two adjacent cameras are located in the same place. However, to avoid mechanical collisions, the mechanism of a given lens assembly, including the sensor package, should generally not protrude outside the frustum of the camera system and enter the conical space of the adjacent lens assemblies. However, the actual lens assemblies in a multi-camera panoramic capture device are also separated by a seam 160. Therefore, if the actual principal rays 170 received at the lens edge, which are inside both the mechanical seam and the physical width or aperture of the mounted outer lens elements (lenses 175 and 180), are projected approximately toward the paraxial NP point 190, they may instead arrive at an offset NP point 192 and be separated by an NP point offset distance 194.

[0038] This can be better understood by considering the enlarged region AA adjacent to the nominal or ideal point NP190, as shown in detail in Figure 5B. Within the hexagonal FOV 182, a ray (e.g., a paraxial ray 173) propagating within the Gaussian or paraxial region and passing through the nominal center of the aperture diaphragm can be projected onto the nominal NP point 190 (corresponding to the entrance pupil) or onto the offset NP point 190A, with a small NP point difference or offset 193 from the nominal NP point 190. On the other hand, the actual hexagonal lens edge principal ray 170 associated with the largest inscribed circle within the hexagon can be projected to reach a common offset NP point 192A, which may have an even larger offset distance (194A). The two adjacent cameras in Figures 5A and 5B may or may not share the same NP point (e.g., 190) located in the same place. Distance offsets can arise from a variety of reasons, including geometric considerations between cameras (adjacent hexagonal and pentagonal cameras), geometric asymmetry within the cameras (e.g., relative to the pentagonal camera), constraints from the practical width of the seam 160, or due to differences in directivity between deviating rays.

[0039] As mentioned above, there are also potential geometric differences in the projection of the incident principal ray toward the simplified nominal "NP point" (190). Firstly, the incident imaging path from near the corners, vertices, or mid-edges (mid-chords) of a hexagonal or pentagonal lens may or may not project to a common NP point within the described range between the nominal paraxial NP point 190 and the offset NP point 192B. Also, as shown in Figure 5B, only due to the geometric asymmetry of the pentagonal lens can the associated pair of edge principal rays 170 and 171 project to different nominal NP points 192B that can be separated from the paraxial NP point (190) by the offset distance 194B and from each other by the offset distance 194C.

[0040] Another issue is that during lens design, the best performance typically occurs on-axis or near-axis (e.g., ≤0.3 field of view (normalized)) and near the optical axis 185. In many lenses, good imaging performance often occurs at or near the edge of the field of view due to the design, in which case optimization weighting is often used to enforce compliance. The worst imaging performance can consequently occur in the middle field of view (e.g., 0.7-0.8 of the normalized imaging field height). Considering Figures 5A and 5B again, although outside the paraxial region, the off-axis rays from the middle field (θ), not as extreme as the edge principal rays (10° < θ < 20.9°), can be projected toward the intermediate NP point between the nominal NP point 190 and the offset NP point 192B. However, other, more extreme, and specifically more affected by aberrations, off-axis rays from the 0.7–0.8 midfield can be projected onto the NP point at a position more or less offset from the nominal NP point 190 than from the offset NP point 192B at the edge of the field of view. To compensate for differences in lens design, the non-paraxial offset "NP" point can be dropped either in front of (closer to the lens) the paraxial NP point (entry pupil) or after (as shown in Figure 2A), as proposed in Figure 5B.

[0041] This is illustrated in more detail in Figure 5C, which essentially illustrates a further zoomed-in area AA of Figure 5B, illustrating the influence of vector projected ray paths associated with deviated imaging rays converging at or near the paraxial entrance pupil (190) for imaging lens systems designed and optimized using the methods of this approach. In Figure 5C, the projected ray paths of green deviated imaging rays in multiple fields of view from the camera lens system converge within a low parallax amount 188 near one or more "NP" points. Similar illustrations of ray fans can be generated for red or blue light as well. The projection of the paraxial ray 173 can converge at or near the nominal paraxial NP point 190, or the entrance pupil located on the nominal optical axis 185 at a distance Z behind the imaging plane 150. The projection of field edge rays 172, including the principal ray 171, converges at or near the offset NP point 192B along the optical axis 185. The NP point 192B can be quantitatively defined, for example, as the center of a large amount of full-field edge rays 172. An alternative offset NP point 192A, corresponding to the “minimum circle of confusion,” can be identified, where the paraxial, edge, and mid- or mid-field rays converge to the minimum spot. These different “NP” points are offset from the paraxial NP point by offset distances 194A and 194B, and from each other by offset distance 194C. Thus, it can be seen that the convergence “NP point” for any given actual imaging lens assembly or camera lens supporting a larger paraxial FOV or asymmetric FOV is typically not a point, but rather an offset low parallax (LP) smudge or quantity 188.

[0042] Within the smudge or low parallax amount 188, various possible optimal or preferred NP points can be identified. For example, the offset NP point corresponding to the field edge ray 172 can be highlighted to help provide improved image tiling. Alternative intermediate field (e.g., 0.6-0.8) NP points (not shown) can also be tracked and optimized. Furthermore, the size and position of the entire “LP” smudge or amount 188, or preferred NP points within it (e.g., 192B), can be modified according to lens design optimization. Such parameters can also vary between lenses for a given design and manufactured lens system due to manufacturing differences between lens assemblies. Figure 5C shows these alternative offset “NP points” 192A,B for non-paraxial rays as being located after the paraxial NP point 190, or further away from the lens and image plane, but other lenses of this type optimized using the methods of this approach can provide similar non-paraxial NP points 192A located in the low parallax amount 188, where they may occur between the image plane and the paraxial NP point.

[0043] Figure 5C also shows the center of the low-parallax multi-camera panoramic capture device, its position relative to the device center 196. Based on optical considerations, the improved panoramic multi-camera capture device 300 can preferably be optimized to nominally position the device center 196 within a low parallax amount 188. The optimized position within this may include being located either or near the offset NP points 192A or 192B, or within an offset distance 194B between them, in order to prioritize parallax control with respect to the principal rays at the edge of the field of view. The actual position there is determined by parallax optimization, which can be determined by lens optimization for spherical aberration of the entrance pupil, direct principal ray constraint, or distortion, or a combination thereof. For example, whether the spherical aberration is optimized to be overcorrected or undercorrected, and what weighting is used for the field operands in the merit function, can affect the positioning of the non-paraxial "NP" point relative to the peripheral or intermediate field of view. The positioning of the "NP" point can also be determined by the control of manufacturing tolerances and residual differences in lens system manufacturing. The device center 196 can also be positioned close to the low parallax point 188, but offset from it by a central offset distance 198. This approach, along with tolerance control, can provide additional space near the device center 196 for cables, circuits, cooling hardware, and related structures. In such a case, the adjacent camera 120 may consequently have an offset low parallax point 188 (Figure 5D) at the “NP” point instead of being in the same location (Figures 5A, 5B). In this example, if the device center 196 were instead positioned at or close to the paraxial entrance pupil, NP point 190, then substantially one or more of the camera 120’s outer lens elements 137 would be smaller than normal, and the desired full FOV would not be achievable.

[0044] Therefore, the no-parallax (NP) point is a useful concept to address, providing useful information for panoramic image acquisition and system design, and assisting in the design of low parallax error lenses, which is ideal, but its constraints also need to be understood. Given this explanation of NP points (multiple) and LP smudge, it is important to understand the ray behavior in this region and define the appropriate parameters or operands to optimize and the appropriate target level of performance to aim for when enabling improved low-parallax multi-camera panoramic acquisition devices (e.g., lens design 320 in Figure 8). In the latter case, for example, a low-parallax lens with a track length of 65-70 mm can be designed for cases where the LP smudge is as wide as 10 mm (e.g., offset distance 194A). However, if this parameter is further improved, alternative lens designs may have a low parallax amount 188 with a longitudinal LP smudge width or a width along the optical axis (offset 194A) of a few millimeters or less.

[0045] The width and position of the low parallax quantity 188, as well as the vector directions of the projections of various principal rays, and their NP point positions within the low parallax quantity, can be controlled during lens optimization by using operands associated with a fan of the principal ray 170 (e.g., Figures 2A, 2B). However, the LP smudge or LP quantity 188 in Figure 5C can also be understood as a visualization of the lateral component of spherical aberration of the entrance pupil, and this parameter can be used in an alternative, but equivalent, design optimization method to the use of the principal ray fan. Specifically, during lens optimization, for example using Code V, the lens designer can create a special user-defined function or operand for the lateral component of spherical aberration of the entrance pupil (e.g., ray height), which can then be used in various ways. For example, the operand value can be calculated as the sum of squared residuals (RSS) of values ​​across the entire FOV or a local field of view, using either uniform or non-uniform weighting for the field of view operands. In the latter case of local field preference, the value can be calculated for the entrance pupil or a position near it, or anywhere within the low parallax amount 188, depending on the preference for the paraxial, intermediate, or peripheral field of view. An equivalent operand could be the width of the minimum circle of confusion in a plane, such as the plane of offset point NP 192A or the plane of offset NP 192B, as shown in Figure 5C. The optimization operand can also be calculated with weightings to reduce or limit parallax error non-uniformly across the field of view, with unbalanced weightings that favor the peripheral or marginal field of view over the intermediate field of view. Alternatively, the optimization operand can be calculated using weightings to provide nominally low parallax error in a nominally uniform manner across the entire field of view (within or across the core FOV 205, as shown in Figure 7). This type of optimization may be particularly useful for mapping-type applications.

[0046] The concept of parallax correction is illustrated in Figure 5D with respect to the center of perspective. The first camera lens 120A collects light from object space 105, including light from two outer ray fans 179A and 179B, whose principal ray projection converges toward the low parallax amount 188A, and images it onto at least one core FOV. These ray fans may correspond to a group 172 of near-field-edge or field-edge rays, as shown in Figure 2B or Figure 5C. As shown in Figure 5C, within the LP amount 188, the vector projection of such rays from object space, generally oriented toward image space, prefers field-edge rays and can therefore cross the optical axis 185 across the imaging plane at or near an alternative NP point 192B, which may be selected or preferred. However, as also shown in Figure 5C, such field-edge rays 172 do not need to cross the optical axis 185 at exactly the same point. When these differences are converted back into object space 105, they translate to small differences in parallax or perspective relative to the imaging ray beam or fan within or across the imaging FOV of the camera lens (e.g., core FOV 205 as in Figure 7).

[0047] As shown in Figure 5D, the second adjacent camera lens 120B provides similar performance and can image the fan of the principal rays 170 from within the core FOV 205, which has a vector projection of these principal rays converging within the corresponding low parallax quantity 188B, including the ray fan 179C. The LP quantities 188A and 188B may overlap, coincide, or be offset depending on the camera geometry and the seam between adjacent cameras, or factors including lens system manufacturing tolerances and compensators, or whether the device center 196 is offset from the LP quantity 188. The more these LP quantities 188 overlap or coincide, the more the perspective centers of the two lens systems overlap. The ray fan 179B of camera lens 120A and the ray fan 179C of camera lens 120B are also nominally parallel to each other, and for example, there is no parallax error between them. However, even if the lens design allows for only very small residual parallax errors at the edge of the field of view, manufacturing differences between lens systems can amplify these differences.

[0048] Analytically, principal ray data from the actual lens can also be expressed as a function of the field of view, in terms of perspective error, including color error. Perspective error can then be analyzed as a positional error in the image between two objects positioned at different distances or directions. Perspective error can be determined by the choice of COP position, the angle within the imaging FOV, and color error. For example, prioritizing the COP to minimize green perspective error may be useful. Perspective difference, or parallax error, can be reduced by optimizing the axial position (Δz) or width of the color within the LP quantity 188, related to the perspective center for one or more field of view angles within the imaging FOV. The perspective center can also be graphed and analyzed as a family of curves, per color, between the Z (axial) intercept position (distance in mm) and the field of view. Alternatively, to get a better idea of ​​how the captured image looks, the COP can be graphed and analyzed as a family of curves relative to the camera system, per color, as parallax error at image pixels, in contrast to the field of view.

[0049] During the design of the camera lens system, the goal may be to limit the parallax error to a few pixels or less for imaging within the core FOV 205 (Figure 7). Alternatively, it may be preferable to limit the parallax error to the outer edge of the core FOV and to the extended FOV region (if provided), particularly in the peripheral field of view. If the residual parallax error for the cameras is therefore sufficiently small, the parallax difference seen as perspective error between two adjacent cameras near their shared seam 160 or in the seam-related region of the extended FOV where imaging overlaps may also be limited to a few pixels or less (e.g., ≤3-4 pixels). Depending on the lens design, device design, and application, it may be possible and preferable to further reduce the parallax error for the lens system to ≤0.5 pixels for the entire core FOV, the peripheral field of view, or both, as measured by the perspective error. If these residual parallax errors for each of the two adjacent cameras are sufficiently small, the image can be acquired, cropped, and easily tiled while correcting or concealing image artifacts from any residual seams 160 or blind areas 165.

[0050] Continuing with the design of that type of panoramic camera in Figure 1, the selection of lens optimization methods and parameters may be important to enable an improved low-parallax multi-camera panoramic capture device (300) with multiple adjacent cameras. A system of camera lenses 120, or lens elements 135, such as that in Figure 2A, can be used as a starting point. The camera lens has compressor lens elements(s) and inner lens elements 140, the latter of which can also be defined as consisting of a wide-angle lens group in front of the aperture and an eyepiece-like lens group behind the aperture. In designing such lenses to reduce parallax errors, it may be useful to consider how the fan of the paraxial-to-nonparaxial principal rays 125 (see Figure 2A), or the fan of the marginal principal rays 170 (see Figure 2B), or the local collection of field edge rays 172 (see Figure 5C) or 179A,B (see Figure 5D) are imaged by the camera lens assembly. While it is possible to optimize lens design by using a set of merit function operands for a set of principal rays (e.g., 31 defined rays), the optimization process can be complex. Alternatively, in tracking the design of an improved low-parallax multi-camera panoramic capture device (300), it was found that improved performance could also be achieved by using a reduced set of ray parameters or operands that emphasize the lateral component of spherical aberration at the entrance pupil or similar selected surfaces or locations within the LP smudge amount 188 behind the lens system (e.g., offset NP points 192A or 192B). Optimization for the lateral component of spherical aberration at the alternative non-paraxial entrance pupil can be achieved using merit function weighting that emphasizes non-paraxial principal rays.

[0051] In another embodiment, in a low parallax multi-camera panoramic capture system, the fan 170 (see Figure 2B) of the principal rays incident on or near the tapered edge of the outer lens element of camera 120 should be parallel to the fan 170 (see Figure 1) of the principal rays incident on or near the tapered surface edge 132 of the outer lens element of an adjacent camera. Note that the “edge” of the outer lens element 137 or compressor lens is a three-dimensional structure (see Figure 2B) and may have a flat edge across the thickness of the glass, and is subject to manufacturing tolerances of its lens element, the entire lens assembly, and the housing 130, as well as adjacent seams 160 and their structure. The positional definition of where the tapered edge is cut into the outer lens element is determined by factors including material properties, front color, distortion, parallax correction, tolerances, and the range of any extra extended FOV 215. The outer lens element 137 becomes a faceted outer lens element when the tapered edge 132 is cut into the lens, nominally creating a set of polygonal edges that follow a polygonal pattern (e.g., a pentagon or a hexagon).

[0052] A camera system 120 having an outer lens element with a polygonal shape that captures incident light from a polygonal field of view results in the formation of a polygonal image on the imaging plane 150, and the shape of the captured polygonal field of view nominally matches the shape of the polygonal outer lens element. Cutting these tapered edges on a given pair of adjacent cameras can affect both imaging and optomechanical structures at or near the intervening seam 160.

[0053] In another aspect, Figure 5E shows “front color,” which is the difference in the nominal ray path between color and field of view, directed off-axis or to the edge field of view. Typically, with respect to a given field of view, the blue ray is offset the furthest. As shown in Figure 5E, the blue ray 157 received on the first lens element 137 is approximately 1 mm further out ΔX than the received red ray 158 directed to the same imaging field of view. If the lens element 137 is not large enough, this blue light may be cut off or blurred, and color unevenness artifacts may occur at or near the edge of the imaging field of view. Front color may appear within the captured image content as a narrow, rainbow-like contour of the polygonal FOV or polygonal edge of the outer compressor lens element 437, which acts as a field of view aperture for the optical system. Local color transfer differences that can cause front color-related color unevenness artifacts near the image edge may arise from differential vinetting at the sloping edge of the outer compressor lens element 137, or from edge truncation at the compressor lens element 438 (Figure 13A), or through the aperture diaphragm 145. During lens design optimization to provide an improved camera lens (320), front color can be reduced (e.g., to a width of ΔX ≤ 0.5 mm) as part of the color correction of the lens design, including by glass selection within the compressor lens group or the overall lens design, or as a trade-off in lateral color correction. The effect of front color on the captured image can also be optomechanically reduced by designing the improved camera lens (320) to have an extended FOV 215 (Figure 7), and by designing the optomechanics to push a straight cut or a sloped lens edge 132 at or beyond the edge of the extended FOV 215, so that residual front color occurs outside the core FOV 220. Front color artifacts can then be removed during the image cropping step in image processing. The effects of front color or lateral color can also be reduced by spatially different color corrections during image processing.Alternatively, an improved camera lens (320) could have a color-dependent aperture, either at or near the aperture diameter, which can provide a larger transmission aperture (diameter) for blue light than for red or green light, for example.

[0054] The optical performance at or near the seam can be understood in part in relation to distortion (Figure 6) and the defined set of fields of view (Figure 7). Specifically, Figure 7 shows the potential set of fields of view that the potential image light can collect by two adjacent cameras. As an example, a camera with a pentagonal outer lens element having a seam 160 separating it from an adjacent lens or camera channel, whether associated with a dodecahedron, truncated icosahedron, or other polygonal lens camera assembly, can image an ideal FOV 200 where the lens extends to the vertex (60) of the frustoconical or conical volume in which it resides, or to the polygonal edge. However, due to various physical constraints that may arise at the seam, including the finite thickness of the lens housing, the physical configuration of the tapered lens element edge, mechanical wedges, and tolerances, a smaller core FOV 205 of the transmitted image light can actually be imaged. The coated aperture for the outer lens element 137 should surround the core FOV 205 with some margin (e.g., 0.5–1.0 mm). Since the lens can be manufactured with an AR coating before the beveling is applied, the coating can extend to the seam. The core FOV 205 can be defined as the largest low-parallax field of view that a given actual camera 120 can image. Equally, the core FOV 205 can be defined as the sub-FOV of the camera channel whose boundary is nominally parallel to the boundary of its polygonal cone (see Figures 5A and 5B). Ideally, with a small seam 160, as well as proper control and calibration of FOV pointing, the nominal core FOV 205 approaches or matches the ideal FOV 200 in size.

[0055] During the camera alignment and calibration process, a series of image references 210 can be established along one or more edges of the core FOV 205 to assist in image processing and image tiling or mosaicing. The resulting gap between the core FOV 205 supported by the first camera and that supported by the adjacent camera may result in a blind area 165 (Figures 5A, 5B). To compensate for the blind area 165 and the associated loss of image content from the scene, the camera may be designed to support an extended FOV 215, which can provide sufficient extra FOV to absorb seam width and tolerances, or offset device center 196. As shown in Figure 7, the extended FOV 215 can be extended sufficiently to provide overlap 127 with the edge of the core FOV 205 of the adjacent camera, although the extended FOV 215 can be much larger. This limited image overlap, as previously mentioned with respect to Figure 3, can result in a small amount of image resolution loss, parallax error, and some complexity in image processing, but it can also help reduce the apparent width of the seams and blind areas. However, as provided by this approach, if the extra overlapping FOV is moderate (e.g., ≤5%) and the residual parallax error within it is sufficiently small (e.g., ≤0.75 pixel perspective error), the image processing burden can be very small. Image acquisition up to the extended FOV 215 can also be used to enable an intermediate acquisition step that supports camera calibration and image correction during the operation of the improved panoramic multi-camera acquisition device 300. Figure 7 shows an inscribed circle within one of the FOV sets corresponding to a subset of the core FOV 205, which is a common core FOV 220 that can be acquired in all directions from that camera. The angular width of the common core FOV 220 can be useful as a quick reference to the image capacity of the camera. An alternative definition of a larger common core FOV 220 may also be useful to include the entire core FOV 205.The dashed line (225), extending from the common core FOV 220 or core FOV 205 beyond the ideal FOV 200 to nominally include an extended FOV of 215, represents a region where careful mapping of principal rays or primary rays or control of spherical aberration of the entrance pupil can be supported in the lens design to enable low parallax error imaging and easy tiling of images captured by adjacent cameras.

[0056] Over a seam 160 extending between two adjacent usable apertures between two adjacent cameras, it may be advantageous for the image light to be captured substantially straight, parallel, and at common intervals with respect to a finite distance in order to reduce parallax and improve image tiling. The amount of FOV overlap required to provide an extended FOV and limit the blind area can be determined by controlling the relative proximity of the entrance pupil (paraxial NP point) or the alternate preferred plane (e.g., to enhance peripheral rays) within a low parallax amount 188 relative to the device center 196 (e.g., the center of the dodecahedron shape). The amount of extended FOV 215 is preferably 5% or less (e.g., an additional field of view of 1.8° or more relative to a nominal core FOV of 37.5°), thereby the peripheral field of view of the camera is consequently, for example, about 0.85 to 1.05. If spacing constraints at the device center and manufacturing tolerances are well managed, the extended FOV 215 can be reduced to an additional field of view of 1% or less. Within the extended FOV 215, parallax should be limited to the nominal system level, while both image resolution and relative contrast remain satisfactory. Parallax optimization to reduce parallax errors can use either principal ray or pupil aberration constraints, aiming for optimization in the high FOV region (e.g., 0.85–1.0 field of view) or beyond, including the extra camera overlap region provided by the extended FOV 215 (e.g., Figure 7, a partial field of view range of approximately 0.85–1.05).

[0057] In addition, controlling image distortion for image light passing through the edges of the outer lens elements' FOV, e.g., in or near the peripheral field of view, can be important when enabling an improved low-parallax multi-camera panoramic capture device (300) with limited parallax error and improved image tiling. In geometrical optics, distortion is a deviation from the preferred state (e.g., linear projection) where straight lines in the scene remain straight in the image. It is a form of optical aberration that describes how rays from the scene are mapped to the image plane. Generally, for image capture, in lens assemblies used for human viewing, it is advantageous to limit image distortion to a maximum of + / - 2%. In current applications, having moderate distortion of 2% or less may also be useful for tiling or combining panoramic images from images captured by adjacent cameras. For reference, in tube distortion, image magnification decreases with distance from the optical axis, and the obvious effect is that of an image mapped around a sphere (or tube). Often used to capture hemispherical or panoramic views, fisheye lenses typically exhibit this type of distortion as a way of mapping an infinitely wide object plane into a finite image area. Fisheye lens distortion (251) can be large as a deviation from f-θ distortion (e.g., 15% or 90° half-width (HW) across the entire field of view), but only a few percent for smaller fields of view (e.g., ≤30°HW). In another example, in laser printing or scanning systems, f-θ imaging lenses are often used to print images with minimal banding artifacts and image processing corrections for pixel placement. Specifically, F-θ lenses are designed with cylindrical distortion that results in linear pixel positioning with a roughly constant spot or pixel size and field of view angle θ (h=f*θ).

[0058] Therefore, an improved low-parallax camera 320 that captures a half-FOV of 35-40° or less may have fisheye distortion 251, as the distortion may be sufficiently low. However, the distortion can be more conveniently optimized for the design of an improved camera lens assembly for use in an improved low-parallax multi-camera panoramic capture device (300). As a first example, it may be convenient to provide a camera lens assembly with local nominal f-θ distortion 250A at or near the edge of the image field, as shown in Figure 6. In one example, the image distortion 250 peaks at about 1% in about 0.75 fields, and the lens design is not optimized to provide f-θ distortion 250 below about 0.85 fields. However, during the lens design process, the merit function can be constrained to provide distortion 250A or a nearly flat distortion 250B for rays imaged at or near the edge of the field, such as for peripheral fields that extend over a partial field range of about 0.9-1.0. This wide field of view range with f-θ type or flattening distortion correction includes the fan 170 of the principal rays or peripheral rays in Figure 2B, including rays imaged through the corners or vertices 60, such as those of a lens assembly with hexagonal or pentagonal outer lens elements 137. Additionally, due to manufacturing tolerances and dynamic effects (e.g., temperature changes) applicable to the camera 120, including both the lens elements 135 and the housing 130, and the collection of cameras 120 in a panoramic multi-camera capture device, it may be advantageous to extend the area of ​​nominal f-θ or flattening distortion in the peripheral field of view beyond the nominal full field of view (e.g., 0.85–1.05). This is shown in Figure 6, in which the area of ​​reduced or flattened distortion extends beyond the full field of view to approximately 1.05 fields of view. In such a peripheral field of view range, it may be advantageous to limit the total distortion change to 0.5% or less. Controlling peripheral field of view distortion keeps the image "edges" straight within adjacent pentagonal regions. This could enable more efficient use of pixels when tiling images, and therefore faster image processing.

[0059] The above explanation discusses distortion in the classical sense, as image aberration at the image plane. However, in low parallax cameras, this residual distortion is typically a trade-off or nominal cancellation of the contribution from the compressor lens element (137, or 437 and 438 in Figure 13A) to that of the collective inner lens element (140, or 440 in Figure 13A). Importantly, the diversion of rays caused by the distortion contribution of the outer compressor lens element also affects both the imaging ray path and the projection principal ray path toward the low parallax amount. This consequently means that, for at least some low parallax lens designs, distortion optimization can affect the optimization of parallax or the field edge NP point or perspective center.

[0060] To support image tiling, the definition of the peripheral field of view or partial field of view range (e.g., approximately 0.85–1.05, or including an additional field of view of 5% or less), where parallax, distortion, relative contrast, resolution, and other performance factors are carefully optimized, can be determined by the device and camera geometry. As an example, the lower limit of the peripheral field of view can be defined as approximately 0.83 for a hexagonal lens and field of view, and approximately 0.8 for a pentagonal lens. Figure 7 illustrates an example with two adjacent pentagonal outer lens elements and FOV sets, but the approach of defining the peripheral field of view and extended FOV to support small areas of overlapping image capture can be applied to multi-camera capture device designs with adjacent pentagonal and hexagonal cameras, or adjacent pentagonal cameras, or cameras with adjacent edges of other polygons or arbitrary shapes or contours in general.

[0061] For the extended FOV 215 to be functionally useful, the nominal image formed on the image sensor corresponding to the core FOV 205 must be underfilled to at least enough to allow the used image area of ​​the image sensor to also be imaged for the extended FOV 215. This may be done to help absorb actual deviations from the ideal of the manufactured lens assembly, or manufacturing differences in the design with offset device center 196 and the improved low parallax multi-camera panoramic capture device (300). However, as will be discussed later, a clever mechanical design of the lens assembly can help reduce parallax errors and FOV overlap or underlap by limiting mechanical displacement or wedge by influencing both the imaging field of a given camera and the seams between cameras. Similarly, compensator or reference adjustment of the image FOV (core FOV 205) size and position, as well as image centroid tracking and shape tracking, may be helpful. Combining distortion optimization and low or zero parallax imaging for extended peripheral vision, careful mechanical design to limit and correct component and assembly differences, and the use of correction criteria or compensators in some combination can provide an excellent overall system solution. As a result, images captured from cameras can be easily cropped to the expected nominal size and shape relative to the nominal core FOV 205, and images from multiple cameras can then be mosaiced or tiled together to form a panoramic image with a reduced burden of image post-processing. However, the extended FOV 215 should, if necessary, provide a sufficient additional angular width (e.g., θ1 ≤ 5% of FOV) to match or exceed the expected wedge or tilt angle q2 that may occur at the seams (θ1 ≥ θ2).

[0062] In designing improved imaging lenses of a type usable in low-parallax panoramic multi-camera capture systems (100 or 300), several primary parameters can be calculated to inform the design efforts. The primary parameter is the target size of the frustoconical or conical volume, based on the selected polygonal configuration (lens size (FOV) and lens shape (e.g., pentagon)) and sensor package size. Other primary parameters that can be estimated include the nominal position of the paraxial entrance pupil, the focal lengths of the compressor lens group and the wide-angle lens group, and the FOV seen by the wide-angle group.

[0063] However, the design optimization for the improved camera lens (320) for use in the improved low-parallax panoramic multi-camera capture device (300) also depends on how a number of other lens attributes and performance metrics are prioritized. Specifically, the relevant system parameters may include controlling parallax or perspective center (COP) errors at the edge or inner field position of the image field, or both, as the optimization is performed using the fan of the principal ray or the spherical aberration of the entrance pupil. These parameters are closely related to other important parameters, including the width and position of the “LP smudge” or quantity 188, the size of any central offset distance between the entrance pupil or LP smudge and the device center 196, the target width of the gap or seam, the range of the blind area 165, and the size of any peripheral or extended FOV to provide overlap. Relevant performance metrics may include image resolution or MTF, distortion (particularly in the peripheral field of view, as well as distortion of the first compressor lens element and compressor lens group), lateral color, relative contrast, front color, and color vignetting, telecentricity, and ghosting. Other relevant design variables may include mechanical and material parameters such as the number of compressor lens elements, the configuration of compressor lens groups, wide-angle lens groups, and fisheye lens groups, glass selection, the maximum allowable size of the first compressor or outer lens element, sensor package size, track length, nominal distance from the image plane to the nearest front lens element (e.g., working distance), nominal distance from the image plane to the entrance pupil, nominal distance from the image plane or entrance pupil to the polygon center or device center, manufacturing tolerances and limits, and the use of a compensator.

[0064] As a second illustrative example, Figure 8 shows a lens design for an objective lens with an alternative improved camera lens 320 or lens element 335, which is an enhanced version of lens 120 of Figure 2A that can be used in an improved low parallax multi-camera panoramic capture device (300). Figure 8 shows a partial enlarged view showing the overall lens shape on the left and the inner lens element 350 in more detail, although Figure 8 does not include an illustration of the lens housing for supporting these lens elements. Also designed for a dodecahedral system, this lens system has a lens element 335 that includes both an outer lens element 345a and a first lens element group or compressor lens group consisting of compressor lens elements 345b and 345c, as well as the inner lens element 350. In this design, the compressor elements 345b,c are not fully bonded as cemented or air-spaced doublets. Similarly, as shown in Figure 8, the inner lens element 350 consists of a front wide-angle lens group 365 and a rear eyepiece-like lens group 367.

[0065] In Figure 8, the lens system of camera 320 collects light rays 310 from object space 305 to provide image light 315 from the field of view 325, directs them through lens elements 335, which consist of outer lens elements 340 and inner lens elements 350, to provide an image at the imaging plane 360. This lens system provides improved image quality, telecentricity, and parallax control, although these improvements are not evident in Figure 8. In this example, the outer lens elements 340 include a group of three compressor lens elements 345a, 345b, and 345c, and the light intensity, or optical bending load, is distributed among the multiple outer lens elements. Image light 310 from object space 305 is refracted and transmitted through the first lens element group or compressor lens group 340 having three lens elements, so that the principal ray at 37.377 degrees at the apex is redirected at a steep angle of about 80 degrees toward the optical axis 385.

[0066] A second lens element group or wide-angle lens element group 365 follows this compressor lens element group, and it consists of two lens elements between the compressor lens element group and the aperture diaphragm 355. A third lens element group or eyepiece lens group 367, having five lens elements, redirects the passing image light coming from the aperture diaphragm 355, providing the image light telecentrically to the image sensor at the imaging plane 360 ​​at F / 2.8. Since this lens is designed for a dodecahedral system, the first lens element 345a nominally receives the image light for an FOV width of 31.717 degrees in the middle chord. The principal ray projection converges or points toward the LP smudge 392, which includes the paraxial entrance pupil.

[0067] As illustrated by Figure 8, this type of camera lens, or lens configuration, comprising a first lens element group or compressor lens group or lens element (345a, b, c), a second lens element group or wide-angle lens group 365 in front of the aperture, and a third lens element group or eyepiece-like lens group 367 behind the aperture, visually resembles a fisheye lens in whole or in part, but it is entirely different. Unlike this lens design (e.g., Figure 8A), a fisheye lens is an ultra-wide-angle lens with severely overcorrected spherical aberration of the pupil, such that its entrance pupil is positioned close to the first lens element and near the front of the lens. This pupil aberration also causes a substantial shift and rotation of the non-paraxial entrance pupil relative to the paraxial entrance pupil. Such a lens also provides a long back focal length by inverting the telephoto aspect ratio, and a positive value for the ratio of the distance from the entrance pupil to the image plane (EPID) divided by the lens focal length (EPID / EFL). Fisheye lenses typically follow a monotonous curve (e.g., H=fθ(f-theta)) and capture images with a characteristic convex, nonlinear appearance, while also providing strong visual distortion. Typical fisheye lenses capture a nominally wide-angle full FOV of 180°, but fisheye lenses capturing images with even larger FOVs (270–310°) have also been described in the literature. In contrast, the improved low-parallax wide-angle camera lens 320 of this approach, used within an improved low-parallax multi-camera panoramic capture device (300), is intentionally designed with low distortion, particularly at or near the edge of the capturing FOV, to facilitate image cropping and tiling. Furthermore, although wide-angle, this camera typically captures image light from a significantly smaller FOV than a fisheye lens. For example, a camera for a regular dodecahedron device nominally captures images from a full-width FOV of approximately 63–75°. On the other hand, an octahedron device can nominally have a camera that captures image light from a full field of view (FOV) of approximately 71 to 110°, and a truncated icosahedron device can nominally have a camera that captures image light from a full field of view (FOV) of approximately 40 to 45°.

[0068] The wide-angle lens group 365 in front of the aperture and the eyepiece lens group 367 behind the aperture are not used as a standalone system for this application if the compressor lens groups 345a, b, c are removed, although these two inner groups can also work together to form an image at or near the image plane or sensor. In the optical design of the camera lens (320), these lens groups, and in particular the wide-angle lens group 365, visually resemble a door peeper lens design. However, although this combination of the two groups of lens elements may again resemble a fisheye or door peeper type lens, they again do not image with fisheye type f-theta lens distortion (e.g., H=fθ).

[0069] In contrast, the rear lens group (367), or sub-system optical structure, is similar in structure to an eyepiece, similar to those used as eyepieces in microscopes or telescopes, but without an eye and used in reverse. An eyepiece is an optical system in which the entrance pupil is always located outside the system. Where the eye may be placed in a visual application, the entrance pupil of the eyepiece nominally overlaps with the plane where the aperture diaphragm 355 is located. Similarly, the nominal input imaging plane in a visual application corresponds to the sensor plane (950) in this application. The eyepiece group (367) is not designed to work with an eye and therefore does not satisfy the requirements of a real eyepiece for interpupillary distance, accommodation, FOV, and pupil size. However, this eyepiece-like lens group solves similar problems and therefore has a shape similar to that of an eyepiece. Depending on the application, the optical design may provide nominal optical performance more or less similar to a more typical eyepiece.

[0070] The improved lens 320 in Figure 8 is similar to the camera lens 120 in Figures 2A,B, but has been designed for a more demanding set of conditions related to the entrance pupil, which has been moved further back to provide further room for parallax correction, an even larger image size (4.3 mm width), and the use of a larger sensor substrate. Since the type of glass can be changed to suitably use both crown and flint type glass, this type of configuration with multiple compressor lens elements may be useful for color correction. In this example, the outer lens element 345a, or the first compressor lens, is a meniscus-shaped lens element of SLAH52 glass with an outer surface 338 having a radius of curvature of approximately 55.8 mm and an inner surface having a radius of curvature of approximately 74.6 mm. Thus, the overall optimized and improved multi-camera capture device 500 may have a nominal radius of approximately 65 mm from the apex of the outer lens element to the nominal NP point position. In this example, the incident light 310 from object space 305, which becomes the image light 315, is significantly refracted inward upon encountering the outer surface 338, but not as dramatically as the refracted inward provided by the first surface of the lens in Figure 2A.

[0071] The requirement for using a larger sensor substrate is to increase the distance between the image sensor surface and the entrance pupil or low parallax 392. Specifically, the focal length is larger (5.64 mm) to project the image onto a larger sensor. Within the LP smudge or low parallax 392, there are several potentially useful reference planes or positions, including the position of the paraxial entrance pupil, or the center of perspective, or the position relative to the non-paraxial principal ray NP point, or the position of the minimum circle of confusion having the minimum size within the plane tangent to the optical axis of the LP smudge or parallax. The entrance pupil is a good reference because it can be easily calculated from a common first-order optical equation. The axial position of the center of perspective is also a good reference because it is directly related to the perceived image quality. The distance from the imaging plane 360 ​​to any of these positions can be used as a reference, but an offset distance 375 relative to the paraxial entrance pupil may be preferred. In this example (Figure 8), with a negative ratio of entrance pupil distance to focal length, EPID / EFL = -5.3:1, the entrance pupil is positioned approximately 30 mm behind the image plane 360. Depending on how it is measured, the LP smudge 392 may have an axial width of ≤2 mm.

[0072] The improved camera lens system 320 in Figure 8 provides an example of how the lens configuration may differ from those shown in Figures 2A and 2B. Generally, lens configurations for enabling improved low-parallax multi-camera panoramic capture devices (300) have a common set of features, consisting of a first compressor lens group that sharply bends light toward the optical axis, a physically much smaller wide-angle lens group that redirects light into the aperture diaphragm, and an eyepiece-like lens group that directs the transmitted image light toward the image plane and focuses it. The requirement to reduce parallax or perspective errors while allowing multiple polygonal cameras to be placed adjacent to each other to form an even larger improved low-parallax multi-camera panoramic capture device (300) leads to extreme lens configurations, in which case the lens elements in the compressor lens group may be somewhat larger (e.g., 80–120 mm in diameter), and more typically, at least some of the lens elements in the wide-angle and eyepiece lens groups may be somewhat smaller at the same time (e.g., 5–10 mm in diameter). In these types of lens designs, the first compressor lens element or the outermost lens element 345a, and adjacent outer lens elements of adjacent lens systems, may alternatively be part of a continuous faceted dome or shell. It is also typical that some (e.g., 2-4) of the lens element surfaces have aspherical or conical contours to bend or direct light rays passing near the edges of the lens element differently from those passing near the center or optical axis. Typically, wide-angle lens group 365 also has lens elements with deeply concave surfaces. In some cases, during optimization, the surface may be desired to be excessively hemispherical, for the purpose of improving element manufacturability, but such contours are preferred to be avoided. Another measurement of the extreme characteristics of this lens form is the offset distance of the paraxial entrance pupil (or similarly, LP smudge) behind or beyond the image plane. Unlike typical lenses, the entrance pupil is not in front of the image plane, but can instead be pushed far behind or beyond it. This is emphasized by the negative ratio of distance from the entrance pupil to the image plane to the focal length, EPID / EFL, which can range from -2:1 to -10:1, but is typically ≥-4:1.

[0073] As illustrated in the illustrative details of Figure 8, optimizing the size, position, and characteristics of the LP smudge or low parallax quantity 392 affects the performance and design of the improved camera lens system 320. Low parallax quantity optimization is largely influenced by the merit function parameters and weighting to the principal ray for both spherical aberration and axial or longitudinal chromatic aberration of the entrance pupil. Lens element and lens barrel manufacturing tolerances can also affect the size and positioning of this quantity, or equivalently, the amount of residual parallax error provided by the lens. Thus, even if these lenses may be considered to have extreme shapes, optimization can help reduce sensitivity to manufacturing errors and provide insight into how and where to provide corrective adjustments or compensators.

[0074] Enhanced situational awareness can be directly enabled through appropriate lens design and the use of optical detectors or sensors by an improved low-parallax multi-camera panoramic capture system (300) equipped with a low-parallax camera lens 300, such as that shown in Figure 8. For example, optical event detection sensors, such as the Oculi SPU, can be positioned on the imaging plane 360 ​​and can use their fast response and large dynamic range to detect sudden changes in objects in the scene. Neuromorphic or event sensor technology is still relatively early in its development, and currently, these sensors tend to have lower spatial resolution compared to CCD or CMOS image sensors. Therefore, as an alternative to provide situational awareness, a high-resolution, large-pixel image sensor, such as the Teledyne Emerald 67M with addressable 67 megapixels, can be positioned on the imaging plane 360 ​​of a well-designed lens 320. However, this sensor is large, and since the camera channel 320 needs to fit within a conical volume or frustum, the front compressor lens elements (345a, b, c) can be very large and difficult to manufacture. These problems can be addressed by reducing the sensor size (for example, in the case of Teledyne Emerald 16M or 36M), by reducing the field of view (FOV) imaged by the camera lens, or by a combination of both. For example, if the entire polygonal shape is changed from a dodecahedron to a truncated icosahedron, the field of view captured by the camera lens (32) decreases, allowing a larger sensor to be supported, improving lens image quality, and resulting in improved angular resolution.

[0075] As an alternative approach that can enable higher resolution imaging or dual modality detection and various situational awareness possibilities, an improved low-parallax multi-camera panoramic capture system (300) may include a low-parallax camera lens 320, which functions as an objective lens, paired with an imaging relay optical system. Figure 9 shows such a system with an objective or camera lens 320, including a compressor lens group 340 paired with an imaging relay 400, where the relay is a lens system with nominal magnification of 1.5x. These lenses are nominally aligned along the optical axis 385. In Figure 9, the example camera lens 320 is similar to that in Figure 8, except the front compressor lens group 340 includes a bonded doublet. In this type of system, the original imaging plane 360 ​​corresponds to the actual spatial image, which is an intermediate image relative to a second imaging plane 410 at the far end of the imaging relay. A large, high-resolution image sensor, such as the Teledyne 67M, can consequently be provided for this second imaging plane 410. The optical system will be appropriately designed so that the optical resolution and sensor resolution are substantially matched. The aperture diaphragm 355 of the objective lens (320) is nominally re-imaged to a secondary aperture diaphragm 455 in the relay optical system. The optical relay design 400 also includes a gap or clearance 420 between the outer surface of the last field lens element 430 and the subsequent lens element. The relay optical system may also include one or more beam splitters for directing light to a secondary optical sensor, such as an IR image sensor or event sensor, provided at the offset or secondary imaging plane. The optical system in Figure 9 can be assembled around a nexus-type internal frame (e.g., Figure 11) that provides a hollow center or open space through which multiple imaging beams of image light from multiple camera channels can intersect with each other.

[0076] Figure 10 then shows an example of an electronic system diagram for a multi-camera capture device 300 of the type shown in Figure 1, where the camera channels 120 are arranged in a dodecahedron shape to directly image each sensor. Image data is collected from each of the 11 cameras 320 and can be directed through an interface input-output module and through cables or cable bundles to a portable computer capable of providing image processing, including live image trimming and splicing or tiling, as well as camera and device control. Output image data can be directed to an image display, a VR headset, or an even more distant computer, either locally or remotely. Power and cooling can also be provided as needed. To help reduce the thermal gradient between the sensors and their electronics and the optical system, microheat pipes or Peltier devices can be used to cool the sensors and redirect the heat. Heat can be removed from the entire device by either active or passive cooling provided through the electromechanical junction at the 12th camera position, as shown in Figure 10. This cooling can be provided by convection or conduction (including liquid cooling) or a combination thereof. External ambient or environmental factors may also affect the performance of the multi-camera capture device 300. This approach can also be used with multi-camera capture devices 300 having other geometric shapes, including those having camera channels with imaging relays (Figure 9).

[0077] Figure 11 shows an example of a mechanical configuration for an internal space frame that can be used in the improved multi-camera acquisition device 300. Specifically, Figure 11 illustrates a mounting assembly that includes a nexus internal frame 500 in which a number of pentagonal faces 510 are arranged in a regular dodecahedron pattern with a hollow center. Generally, the internal frame 500 is a polygonal frame having mounting and alignment features, with an array of adjacent mechanical faces and peripheral edges forming a polygon. The internal frame 500 can be designed as a mounting or mechanical assembly for supporting 11 camera systems with a support column mounted in a 12th position (similar to Figure 10). Polygonal internal frames, or half or partial internal frames, can also be used in partial or hemispherical systems, in which case camera assemblies, including image sensors, are mounted directly or indirectly to the frame. Connections, cables, and wiring for data transfer and cooling are then directed outward through an open polygonal portion 530 of face 510 and enter the hollow center of the internal frame 500, and can exit through an open polygonal portion 530 of another face 510. Alternatively, a hemispherical system with an internal mounting frame 500 can provide a hollow center or open space (e.g., a nexus) to allow an image light beam to intersect through a pair of opposing open polygonal portions 530 of face 510 so that it passes through a subsequent relay optical system (400) and reaches a distant optical sensor in the secondary imaging plane 410. Positionally, the width of the gap or clearance 420 in the relay optical system (see Figure 9) between the outer surface of the last field lens element 430 and the nearest subsequent lens element 435 nominally coincides with the width of the central hollow volume between the opposing faces 510 provided by the nexus internal frame 500. For example, the clearance 420 may be 75 mm wide. However, it should be noted that the objective lens housing or relay lens field lens elements 430 and their housings may protrude moderately through the open polygonal portion 530 of the surface 510 and enter the central volume of the hollow center 540, provided that they do not obstruct the imaging light of the adjacent objective lens 320. In such a case, the clearance between the lens elements will be less than the width of the hollow center of the internal frame 800.For example, the width of clearance 420 could be a few millimeters, which is smaller than the center width of hollow center 540.

[0078] As shown in Figure 11, the nexus internal frame 500 may have a pentagonal face (510A) which may have three adjusters 520, such as positioning screws or flexures, nominally oriented at 120° apart, that interact with mounting and alignment features on the camera housing and can therefore be used to assist in aligning a given camera channel. For an improved multi-camera panoramic image acquisition device 300 constructed with a dodecahedron pattern, the internal frame may also be a dodecahedron having a pentagonal face, which may be oriented with an internal pentagonal face nominally aligned with an external pentagonal geometry. The internal spatial frame approach can be used with other polygonal device structures, such as those for octahedrons, icosahedrons, or chamfered dodecahedrons. In such cases, at least a portion of the spatial frame faces may include edges along their periphery corresponding to other polygonal shapes, such as hexagons.

[0079] The internal frame 500 may be machined separately and assembled from two or more pieces that are mounted together, or it may be made as a single-piece molded structure by casting or 3D printing. While the manufacture of a single-piece molded frame is more complex, the resulting structure is more rigid and robust and can support tighter machine tolerances. For example, a dodecahedral frame (500) with a hollow center may be cast in stainless steel and then selectively post-machined on a face 510 to provide precision datum features, including flat sections, V-shaped slots, or ball mounting features. Specifically, one or more pentagonal faces 510A, 510B, or 510C may have one or more adjusters 520 that can be used to lightly press each camera channel against a precision V-groove structure (not shown in Figure 11). These V-groove structures may be made on the inner edge of the pentagonal vertices 60 of the pentagonal face, or protrude from the inner edge. Alignment balls may be mounted on the face 510 or on adjacent lens housings that are joined together, or on a combination thereof. Various features, including balls, V-shapes, flat sections, and sockets, can be used to enable kinematic constraints between lens housings or between lens housings and the spatial frame 500. This internal frame 500 can then be provided with flexures or adjusters on all or most of its pentagonal surfaces to provide kinematic type adjustments, thereby reducing or avoiding excessive constraints during device assembly and use.

[0080] As previously mentioned, the mounting and adjustment for secondary channels may have a different design or configuration than that for primary channels. In these improved devices (300), springs, flexures, magnets, or adhesives can be used on or within the internal frame 500 to provide low-stress mechanical coupling or connection between adjacent camera channel lens housings, and between camera channels and the nexus internal frame 500, or between different parts of the internal frame, thereby helping to limit under-constraint or over-constraint between assemblies or lens housings. Alternatively, the internal frame can be made, at least in part, from a more compliant material such as brass or Invar. 3D printed frames can be manufactured from materials including plastic, bronze, or steel.

[0081] The internal spatial frame for the improved multi-camera capture device 300 can also be a kinematic structure, in which the individual faces (510) are attached to each other using kinematic features. While the resulting spatial frame structure may not be very rigid, it may be easier and cheaper to manufacture and assemble the individual faces than to machine or cast a whole or one-piece spatial frame. However, in that case, the assembled spatial frame should be a kinematic structure so that it can compensate or correct for external loads. The kinematic joints between spatial frame faces can also serve the kinematic interactions between adjacent camera channels attached to the spatial frame. A precisely constrained or kinematic structure (these terms can be used interchangeably) also avoids stress and deformation when assembled in the face of manufacturing variations. It also exhibits precision, returning to a consistent position after experiencing a certain temperature change. These properties make it ideal for an optical support structure. The number of kinematic elements can be based on the system requirements.

[0082] Figure 12A shows an example of a precisely constrained spatial frame structure or kinematic spatial frame 600 of a dodecahedron, where each pentagonal face 610 (or facet or lateral surface) is an independent, precisely constrained element. Each face 610 has a nominally straight edge along its periphery, forming a polygon, which in this example is a pentagon. In other examples, the edge may not be straight and / or there may be more or fewer edges. Figure 12A is illustrated to show key features that enable the kinematics of the spatial frame 600. Each pentagonal face 610 includes 10 mounting points 690, e.g., two for every polygonal edge 680. The constrained face 610 is provided with a specific set of kinematic elements 650 in a subset of its 10 common mounting points 690, and the type of kinematic elements 650 used is determined by the position of the face 610 and its associated mounting points 610 within the overall structure (spatial frame 600). In Figure 12A, a portion of the surface 610 is shown with kinematic elements 650 at mounting points 690, while a portion is not shown, allowing the mounting points to be seen. These mounting points 690 are shown as rectangular cutouts. The mounting points 690 can be located anywhere along the edge 680, but positioning them near the vertices 60 may improve structural stability. The kinematic elements 650 provide a kinematic joint or connection across the seam 615 between adjacent mounting points 690 of two adjacent surfaces 610. The combination of kinematic elements 650 arranged around the spatial frame 600 serves to prevent or reduce over-restraint within the assembly. Figure 12A shows an example of a kinematic spatial frame structure with a surface 610 containing an open polygonal center 630 or opening through which a camera channel lens housing (not shown) can be mounted or inserted. However, additional examples of features on the pentagonal surface 610 for mounting or aligning the camera channel lens housing are not shown.

[0083] A kinematic element 650 mounted at point 690 can appropriately orient itself at an angle between faces 610 to eliminate a different number of degrees of freedom (DOF). Embodiments of these types of kinematic elements 650 are shown. In Figure 12B, a kinematic element 650A, which can be used to eliminate a single degree of freedom, is shown in both perspective and side view, for example, as an example of a kinematic element 650. This kinematic element 650A is embodied as a sphere on a flat portion. It includes two components: a ball mount 652 with a ball 655 or partial ball, and a flat mount 657 with a flat portion 660. A feature 662 for providing a holding force or constraint vector 664 is shown only in part in this figure. Balls or partial balls for use in this type of kinematic element can be obtained from Bal-tec (Los Angeles, California, USA).

[0084] Figure 12C shows a second kinematic element 650B, which can be used to eliminate two degrees of freedom, as an example of kinematic element 650, in both perspective and side views. In this example, kinematic element 650B is embodied as a sphere in a V-shaped socket, which includes two components: a ball mount 652 with a single ball 655 or a single partial ball, and a V-shaped mount 666 with a V-shape 668. In this figure, both components, the ball and the flat mount, include a hole 662. Mechanisms such as springs (not shown) can provide a retaining force along a constraint vector 664.

[0085] In Figure 12D, a third kinematic element 650C is shown in perspective, illustrating two adjacent components that can be used to eliminate three degrees of freedom. In this example, the kinematic element 650C is embodied as a sphere in a three-sided socket, which includes two components: a ball mount 652 with a single ball 655 or a partial ball, and a socket mount 670 with a socket 672. Again, a hole feature example (662) is provided for mounting a retaining force mechanism. This kinematic element can operate like a spherical joint or a Heim joint.

[0086] In the kinematic space frame 600, substitutes for these kinematic elements 650 can be employed, consisting of alternative elements such as wires, blades or V-shaped flexures or combinations of ball and socket joints, working only to eliminate the indicated degrees of freedom between the attached elements. The sum of the constraint vectors 664 nominally coincides with the bisecting angle between adjacent faces 610. With respect to the shown kinematic element example 650, it is assumed that nesting or holding forces can be applied to keep the surfaces in contact and prevent under-constraint. These forces, represented by the constraint vectors 664, can be achieved using force elements such as retaining springs (not shown) attached to holes 662. Alternatively, or additionally, holding forces can be provided by magnets, elastics, adhesives, gravity, or by actively applied external forces such as piezoelectric elements, solenoids, or air or electric cylinders (pneumatic), or combinations thereof.

[0087] There are several ways to create the required constraint patterns for a precisely constrained configuration using these elements. While any face 610 of the spatial frame 600 can be considered fixed, for the sake of this explanatory consideration, the top face is assumed to be fixed to the ground (e.g., to a stable mounting surface during manufacturing). Figure 12E shows a partial spatial frame 600 on the left, with a top face 611 and one of the adjacent faces 610 facing downward at a suitable angle to the dodecahedral structure. Two kinematic elements 650 are shown connecting the faces 610 across a seam 615. In this example, the kinematic element 650 on the left is of type 650C, which removes 3 degrees of freedom (DOF), and the kinematic element on the right is of type 650B, which removes 2 DOF. The resulting constraint pattern allows for a single degree of freedom of rotation around a line (shown as a dashed line) connecting the centers of the two spheres. On the right, Figure 12E shows a more complete partial spatial frame 600 having a left-to-right pair of kinematic elements 650C and 650B connecting the front faces, as provided, by attaching an additional adjacent face 610 with the same constraint pattern to the top face 611 or the ground. Two constraints here eliminate two rotational degrees of freedom (one for each face 610, with the top face 611 fixed to the ground). Ball and V-shaped (Figure 12C) kinematic elements 650B are provided within the lower joint 615 to eliminate 2DOF and connect the left and right faces 610 together. The result is a rigid structure. The remainder of the hemispherical portion of this dodecahedral spatial frame example 500 is then constructed on this partial structure.

[0088] Figure 12F shows a further assembly of the partial kinematic space frame 600, including the next face 610 added to this hemispherical portion. The three previously added faces 610, described in Figure 12E, result in a rigid structure, while the remaining faces 610 within the hemispherical portion can be constrained differently. In the lower right, the added face 610 is constrained to the upper face 611, which is fixed to the ground, using two elements of type 650B in a V-shaped (668) orientation aligned in the direction of the intersection of the two faces 610 and 611. This allows for rotation around the dashed line shown and translation along that line. Again, ball and V-shaped (Figure 12C) kinematic elements 650B to eliminate 2DOF are provided at the lower joint 615 to connect the right face 610 to the left pair of faces. This kinematic element of style 650B between the previous rigid structure and this new face becomes an additional rigid element, eliminating an additional 2 degrees of freedom. Surface 610 can be added in a similar manner around the upper surface 611, completing the hemispherical portion of the partial kinematic space frame 600.

[0089] Alternatively, a similar rigid structure to the partial kinematic space frame 600 can be achieved by having two initial faces (single-degree-of-freedom faces) at the end and beginning of the chain of faces 610. Similarly, the single-degree-of-freedom faces can be positioned across faces with four degrees of freedom, resulting in the same rigid kinematic structure.

[0090] Figure 12G then shows two hemispherical or partial kinematic space frames coupled to a single dodecahedral kinematic space frame 600. The two hemispheres are connected by a connection, which is a "Maxwell" or 2-2-2 type kinematic connection. In principle, a 2-2-2 connection consists of three spheres or balls in three V-shapes spaced roughly 120 degrees apart. A portion of a 2-2-2 connection 675 is shown in Figure 12G, where the space frame 600 has two single-degree-of-freedom kinematic elements 650A (e.g., Figure 12B) on either side of its vertices, connecting the upper and lower hemispherical portions of the space frame together. These two nearly adjacent single-degree-of-freedom kinematic elements 650A effectively function as a double-degree-of-freedom kinematic element 650B. Note that the connection in this example is generated by a simple 90-degree rotation of the planar element around a line roughly passing through the center of the sphere located in a plane perpendicular to both faces 610 and in the plane of the face with the planar element fasteners. Two other kinematic connections, which are 2-degree-of-freedom V-groove type kinematic elements 650B, are located opposite the spatial frame and are not shown as they are not visible. Those two kinematic elements 650B in the style of Figure 12C can be mounted from the upper side of two adjacent faces 610 to near the lower hemispherical vertex 60. The orientation of the three kinematic elements is largely symmetrical to minimize dispersion associated with thermal movement. Alternatively, these two kinematic elements 650B can be mounted at two mounting points (690) of a single face 610, potentially sacrificing some kinematic purity for increased robustness.

[0091] The size, assembly considerations, and / or requirements of this system will determine modifications or adaptations in the constraint pattern. Figure 13A shows an alternative example of the dodecahedron kinematic space frame 700, with a crown-shaped "non-hollow" upper hemispherical portion 720 and an arrangement of six fixed continuous surfaces 730 with a rigid joint or continuous seam 745. The crown 715 may be machined from bulk material, fabricated by additive manufacturing or similar. The lower hemispherical portion consists of discretized surfaces 710 connected by kinematic elements 650, such as kinematic elements 650A, B, C, as previously presented. The crown structure 715 within the upper hemispherical portion 720 may be useful in joining with other system mechanisms requiring greater robustness or rigidity. Lens housings can be mounted on both the fixed surfaces 730 and the kinematic surfaces 710.

[0092] However, the crown structure 715 is more likely to be used as a mechanical connection to other devices or structures, including a lattice structure (not shown) capable of supporting an imaging relay system (Figure 9). Alternatively, the crown structure 715 may have a set of outwardly extending portions that form a cylindrical base and a point at the apex. This could be the alternative upper structure in Figure 13A, while the lower hemisphere still includes a discretization surface 710 with kinematic elements 650.

[0093] As another alternative kinematic space frame 700, Figure 13B shows an upper hemispherical portion 720 including a “fixed” section 725 having two adjacent non-hollow or continuous polygonal faces 740 with a rigid joint or continuous seam 745. For example, the second 725 may be machined from bulk material. In this example, the upper hemisphere is divided into three identical sections, each having two faces 740. One fixed section 725 includes the top surface, and the other fixed sections use a 2-2-2 type connection and are connected to it using an array of kinematic elements 650 of that type (650B in Figure 12C). Thus, the two fixed sections 725 are interconnected using three spheres in three V-shapes. Again, the lower hemispherical portion consists of discretized faces 710 interconnected by kinematic elements 650, such as kinematic elements 650A, B, C, etc., as previously presented.

[0094] Figure 13C shows a third alternative example, where the spatial frame 700 consists of two hemispherical crown portions 715A and 715B, each machined, cast, or printed to have a set of continuous surfaces 740. A kinematic mounting using a suitable set of kinematic elements 650 positions the two semi-crowns relative to each other to provide the complete spatial frame 700. When the two semi-crowns are separated, access to the fasteners that hold the camera lens housing to the spatial frame is relatively easy. The two semi-crowns can be joined together with fasteners accessed through an opening at the bottom. Each crown 715 may be a selectively externally machined, integrally molded cast part. Generally, the inner surfaces of the crowns do not require precise features. However, for an improved camera device 300 with crown-type spatial frame portions (e.g., Figures 13A and 13C) requiring high-precision alignment tolerances, a 5- or 6-axis milling machine may be required to manufacture these structures.

[0095] With respect to the alternative spatial frame examples in Figures 13A, 13B, and 13C, the upper and lower hemispherical portions of the spatial frame 700 are interconnected using a suitable arrangement of kinematic elements 650 (e.g., kinematic elements 650A or 650B). These alternative spatial frames 700 may be useful in reducing the number of individual components and improving the overall robustness of the spatial frame. Other combinations are also possible. For example, a crown structure 715, as shown in the upper hemisphere 720 of Figure 13A, can be joined using the approach shown in the upper hemisphere of Figure 13B, which uses an intermediate number of continuous faces 725. These types of spatial frames, which partially combine kinematically connected discrete non-hollow faces, can be extended to other polyhedral shapes, such as truncated icosahedrons (soccer balls) or chamfered dodecahedrons.

[0096] Figure 14A shows another alternative example 800 of the kinematic space frame to those in Figures 12A and 13A, 13B, and 13C, again having a dodecahedral shape with 12 faces or facets. Face 810 may include a channel V-shape 815 to which a camera channel lens housing (not shown) can be attached. Figure 14B shows an example of the assembly process for a plate or face 810 that can be used in the space frame of Figure 14A. A series of retaining pins 820 are added to face 810 by inserting them into drilled holes and attaching them to the available ends of the pins 820 using springs 825 fixed within face 810. Subsequently, cylinder pins 830 similar to rotary pins are added and pulled by springs 825 against a pair of machined V-shaped grooves 835 provided on the edge 840. Each spring extends from the retaining pin 820 to the narrow end of the cylinder pin 830 similar to a rotary pin. To resist bending under load, these cylinder pins 830 can be made of steel.

[0097] This same assembly process is then repeated for all 12 space frames 810 and their attached pins, springs, and cylinders, however, not all edges 840 of all faces 810 are initially fitted with cylinder pins 830. The individual faces 810 are then joined together, as shown in Figure 14C, to form a space frame 800 of the type shown in Figure 14A. One face 811, and the corresponding lens channel opposite the mounting plate or face 812, can be identified as the primary channel. The secondary faces 810 are joined to the primary faces using springs, which are located on the polygonal edges of two adjacent faces 810 that are joined to the reduced or diminished radius portion of the rotating pin cylinder 830. Once the five pentagonal secondary faces 810 are joined to the primary faces 811, they can be joined to each other. The tertiary faces can be joined to the secondary faces, then to each other, and then to the mounting plate face 812, completing the space frame. In this design, the offset V-shaped groove 835, which is in contact with the cylinder pin 830 and the two adjacent edges 840 of the adjacent surface 810, acts or functions as a kinematic element. The contact of the cylinder with the offset V-shape prevents or resists over-confinement, and the pair of attached springs 825 prevents or resists under-confinement. Thus, this kinematic element using the cylinder pin and pair or the offset V-shape or V-shaped groove is an example of a kinematic element different from the kinematic elements 650A, B, and C described above.

[0098] The spatial frame 800 in Figure 14A provides a kinematic assembly element with spring-loaded connections, but it also includes redundant constraints for cylinder and V-shaped connections along a common edge 840. As a result, the system's rigidity is significantly increased, and the assembly method is common to all edges 840, simplifying overall manufacturing. However, an intermediate design to that of Figure 14A, using a combination of the kinematic element's cylinder pin 830 and kinematic element 650 (e.g., 650A, 650B, or 650C) positioned on one or different edges 840 of the polygonal face 810, can also be used to provide improved or different kinematics between two or more polygonal faces 810. This system also creates a large internal hollow space, allowing for easy complete or partial disassembly for field replacement of components. Again, this spatial frame construction is not limited to a dodecahedral shape; it can also be constructed using a combination of discrete, robust, continuous faces.

[0099] As an alternative to the low-cost version of the spatial frame 800 in Figure 14A, the individual rotating pin cylinders 830 can be replaced with rigid wires wound around edges 840 that form the periphery of the polygonal surface 810. In this example, the alternative surface 810 may have these wires. The spatial frame would consequently have lower manufacturing costs, but some mechanical precision would be sacrificed.

[0100] As shown in Figure 11, the internal space frame 500 can be a rigid structure with machined, cast, or printed continuous surfaces 510. Alternatively, a kinematic space frame 600, as shown in Figures 12A to 12G, can be provided using a set of discretized surfaces 610 and joined kinematic elements 650. Alternative kinematic or partially kinematic space frames 700 can also be provided, having a combination of discrete and non-hollow or continuous surfaces, as shown in Figures 13A, B, and C. Another alternative example of a space frame 800, shown in Figure 14A, may also have a partially kinematic structure. The relative kinematic performance of these spatial frame examples varies due to design choices made in the design and the selection of kinematic elements used, the use of a complete set of discretized polygonal spatial frame faces (e.g., Figure 12A or Figure 14A) versus the use of a structure with 2-3 consecutive polygonal faces (e.g., Figure 13B) versus the use of a structure with one or more larger components with several consecutive polygonal faces (e.g., the crown structure in Figures 13A, C). Relative kinematic performance may also be determined by other factors, such as the material from which the polygonal faces or kinematic elements are manufactured (e.g., steel, aluminum, zinc, bronze, Invar, or FRP). Assembled spatial frames can be pre-tested for kinematic performance. For example, a high-precision kinematic spatial frame can control the relative positioning of one face to another to a spatial tolerance of approximately ±0.025.

[0101] As shown in the figures, the spatial frames 600 in Figures 12A and 12G are nominally fully kinematic due to the appropriate selection of kinematic elements 650A, B, and C between the various faces 610. In particular, the spatial frame 800 in Figure 14A has cylinder pin (830)-based kinematic elements positioned within all joints between all polygonal edges 840, but this spatial frame 800 is either over-constrained or only partially kinematic. However, the kinematics can be improved by replacing some of the cylinder pin (830)-based kinematic elements with alternative kinematic elements, such as using an appropriate selection of kinematic elements 650A, B, and C (see Figures 12B to 12D) between different joints.

[0102] In comparison, the spatial frame 700 in Figure 13B includes both adjacent surfaces 710 that are continuous or rigid rather than hollow, and many other open joints where adjacent surfaces are connected by kinematic elements 650 (e.g., 650A, 650B, or 650C). This frame can be made fully kinematic with appropriate selection of kinematic elements to span the open joints. However, the joints between adjacent continuous surfaces are also nominally rigid since these surfaces are manufactured (e.g., machined, cast, or printed) as a continuous material (e.g., steel). The rigidity of these polyhedral structures can be enhanced by manufacturing surfaces with reinforcing ridges. Alternatively, the rigidity can be reduced by manufacturing these continuous surfaces from one or more compliant materials (e.g., polymers or plastics). Similarly, the spatial frame 700 in Figures 13A and 13C, including the crown-shaped portion, is only partially kinematic, but the open seams can be kinematically bridged, for example, to provide a fully kinematic lower hemisphere (e.g., Figure 13A).

[0103] Therefore, a partially kinematic space frame may have kinematic elements at all joints between adjacent faces, but it does not have a proper selection of kinematic elements across all joints to provide a fully kinematic structure. The resulting structure may be either over-constrained or under-constrained. A partially kinematic space frame may, as an alternative or similar, have some joints between rigid adjacent faces, for example, since the faces are manufactured from a single continuous material. In the examples of this disclosure, an element or body is "fully kinematic" if it is precisely constrained, with six degrees of freedom (DOF) intentionally and appropriately removed, and without redundant or under-constraints (e.g., the element or body is partially unconstrained). For a space frame, this means that individual elements or bodies (e.g., a structure with individual faces or multiple continuous faces) are precisely constrained within the joints between these bodies in order to function or behave as a rigid structure in the composite.

[0104] Any of these kinematic spatial frame examples, or any variation thereof, can be applied to enable an improved panoramic multi-camera capture device 300 with enhanced kinematic performance to help maintain structural integrity during potentially excessive or under-constrained conditions (e.g., changes in load forces, gravity, and thermal conditions), and thus to help hold the camera channels (320) properly aligned with each other. In addition, applying any of these spatial frame examples to the improved panoramic multi-camera capture device 300 for the purpose of supporting the camera channels 320 may also involve providing a kinematic junction between the spatial frame plane and the lens housing or camera channels.

[0105] Furthermore, defining a primary plane is useful in the design of a spatial frame. Similarly, defining a primary camera channel corresponding to a primary plane can be useful. In some designs, the mounting and adjustment of secondary or tertiary camera channels may have a different design or configuration than that of the primary camera channel.

[0106] Figures 15A to 15C illustrate a preferred approach for mounting lens housings to faces or facets of a spatial frame using kinematic balls and V-shaped contact surfaces or features. In this design, nominally all camera channels and their lens housings 905 are similarly mounted to polygonal spatial frame facets 910. For example, they can be mounted to the lens housings 905 with Belleville washers and 4-40 bolts passing through facets 910 of the spatial frame 900, although nesting forces (e.g., 20N) can be provided by other means. The interaction of the balls 930 with respect to the V-shapes 940 provides kinematic positioning that limits constraint problems. Specifically, each ball 930 is located within the V-shape 940, which may eliminate two degrees of freedom in each ball 930, one on each face of the V-shape 940. Since the ball and V-shaped groove features are located on the underside of the lens housing 905 rather than within the seam 902 between adjacent camera channels, it is easier to limit the seam width (e.g., ≤1 mm).

[0107] Figure 15D shows a second example of mounting a camera channel 920 with a lens housing 905 to a surface 910 or facet of a spatial frame 900. In this figure, the spatial frame portion example can be a single-piece crown structure as shown, but it can also be a structure with multiple kinematically mounted parts (e.g., Figures 12A, 14A, or 13A, B). The camera channel 920 with the lens housing 905 shows the protruding shape of the outer compressor lens element 925. The surface 910 of the spatial frame 900 includes three channel V-shaped slots or V-shapes 940 used to kinematically align the surface to the lens housing 905. The lens housing 905 also has three precision balls 940 used to precisely align its assembly to the V-shapes 940 of adjacent surfaces 910. Three fasteners (not shown), such as bolts or magnets, consequently provide retaining force for holding the lens housing 905 to the surface 910 of the spatial frame 900.

[0108] As an alternative to the ball 930, three pins (not shown), mounted on the underside of the lens channel housing 905, contact three corresponding V-shapes within facet 910 of the spatial frame. The pin positions on the lens housing are spaced 120° apart. Each pin is perpendicular to the lower edge of the lens channel on the plane in which it is positioned. Nesting forces can be achieved by placing a cantilever directly over each pin. To eliminate ambiguity regarding the precise position of the pin / V-shape contact surfaces, the V-shapes may be slightly curved to ensure point contact. Compared to the ball and V-shape approach, which creates point contact, the pins would provide line contact, which can result in much lower mechanical stress. However, the protruding pins occupy more space, which may be difficult to provide for this application.

[0109] In either case, it is desirable to minimize the force range at the six contact points of the camera channel. This ensures positive contact forces on all joints without exceeding the yield stress. The design approach shown in Figures 15A–15C can improve performance compared to other approaches because the moment arm is reduced as the mechanical contacts move closer to the center of mass.

[0110] To ensure that the designed constraint pattern between the lens housing 920 and the spatial frame (e.g., 500, 600, 700, or 800) provides accurate constraint, the design was evaluated against Maxwell's Criterion for Accurate Constraint (Douglass L. Blanding, 1995). In this design example, the lens housing contacts associated spatial frame polygonal facets, located within a V-shaped groove and spaced 120° apart, using pins. The lens housing 905 can be bolted to the associated spatial frame polygonal facets, which prevents under-constraint between the two.

[0111] Over-constraints between the lens housing and the associated polygonal facets can also be checked using Maxwell's Criteria. Since Maxwell does not describe cases for five or six constraints, these criteria are designed for four constraints; however, they can nevertheless be used to assess over-constraints by considering three initial constraints, ensuring that each of the remaining constraints, if added, does not violate Maxwell's Criteria. Since all six constraints are sets of three coplanar constraint pairs, no four are on the same plane, and therefore the first criterion is satisfied. Furthermore, the maximum number of constraints intersecting at any given point is two, and therefore the second criterion is also satisfied. By inspection, it is clear that no two constraints are parallel, and therefore Maxwell's third check is passed. The final criterion stipulates that the fourth constraint must not belong to the same set of generators of a single sheet of hyperbolas as the first three constraints. If we identify the first three skew constraints as protruding from the same isosurface of each V-shape, then it is clear that each of the remaining constraints intersects two of the original skew lines, and therefore each of the remaining constraints is not part of this set. Thus, it must lie on the reciprocal generator of the hyperboloid, and thus Maxwell's fourth and final criterion is satisfied. One last remaining check is to ensure that the constraint pattern is not affected by thermal expansion. To do this, it is observed that each constraint surface is arranged radially from the origin, which is defined as the center of the spatial frame surface. Since they are all arranged on radial paths, the expansion is the same proportional distance from the origin for all constraints, the center does not change, and therefore the kinematic coupling is thermally stable.

[0112] Figure 15B then shows two adjacent lens channel housings 905 (lens elements not shown) mounted on two adjacent surfaces 910 of the spatial frame. To enable the intended optical performance, the width or gap of the mechanical seam 902 between adjacent camera channels 920 should be small (e.g., <1 mm) while remaining robust and avoiding interference during use. Any small geometric changes at the spatial frame level, such as changes in V-angle, V-dimension, edge length, pin size, and ball size, will have a significant impact on the seam width. Figure 15C then shows a fully assembled spatial frame with the array of lens housings 905 mounted, enabling the improved panoramic multi-camera capture device 300 (note: lens elements and other components not shown). During the assembly of the device, it is necessary to mount one or more of the lens housings 905, or other components such as an image sensor or data path or power electronics or cooling hardware. This can be done using a variety of features and tools, including a tool with a flexible cable extender.

[0113] As mentioned above, camera channels can also be mounted to the spatial frame using Belleville washers and 4-40 bolts that pass through the facets of the spatial frame to attach them to the lens housing. However, the order of assembly may be important. For example, tertiary camera channels can be mounted first, with their 4-40 bolts accessible from the secondary channel surface. Secondary camera channels adjacent to primary camera channels can then be added, with their 4-40 bolts accessible from the top or primary surface, using flexible drives. Primary or top channels can be added last, with their 4-40 bolts accessible from the base plate if the associated mounting points can be removed.

[0114] It should be noted that balls, V-shapes, and pins should all be surface-hardened to avoid structural failure. While it may be convenient to perform surface hardening on the steel used to construct the spatial frame faces, due to strict tolerance requirements and the high cost of hardening small localized areas, it may be more beneficial to design recesses within the spatial frame faces and then install pre-hardened V-shapes. Such V-shapes are readily available from suppliers such as Bal-tec and can be easily incorporated into spatial frame designs.

[0115] In some of the exemplary embodiments, springs are shown to connect the surfaces of the spatial frame together, but other retaining mechanisms, including magnets, can be used. Specifically, magnets can be used to assist in holding the spatial frame together and / or to hold the lens housing to the spatial frame surface. As an example, a permanent rare-earth magnet, part number D32SH, with a diameter of 3 / 16 inch × thickness of 1 / 8 inch and a tensile force of 1 to 2 pounds, from K&J Magnetics of Pipersville, Pennsylvania, USA, can be used. With a close gap of 0.75 mm between two opposing magnets, the attractive force between the two magnets may be approximately 0.5 pounds.

[0116] Furthermore, as shown in Figure 15D, during assembly, the lens housing 905 is positioned with its mounting surface 935 very close to the outer surface 912 of the surface 910, and fits into the opening 915. Precise mounting is determined by the interaction of the ball 930 with the V-shape 940. The ball 930 and the V-shape 940 are nominally spaced 120 degrees apart. As shown in the figure, the mounting surface 935 is located approximately midway along the length of the lens housing 905. However, alternatively, the mounting surface 935 can be provided closer to the position of the imaging plane or image sensor, or closer to the position of the outer compressor lens element 925. A preferred approach is to position the mounting surface 935 near the center of mass or center of gravity of the camera channel.

[0117] Similarly, as shown in Figure 15D, the lens housing 905 of the camera channel 920 may include magnets 945, balls 950, and flat sections 955 on the side walls of the lens housing. These features can provide kinematic constraints between adjacent camera channels. Specifically, the magnets can be used to assist in the kinematic mounting and alignment of the first camera channel to the second adjacent camera channel by providing a load force. The use of such side wall magnets (or latches or other mechanisms) may be of greater value if the mounting surface 935 is closer to the image plane or closer to the outer compressor lens, rather than closer to the center of the lens housing (as shown in Figure 15D) or the center of mass of the camera channel. In such cases where the mounting point is at or near an extreme mechanical position relative to the length of the camera channel or objective lens assembly, the side wall kinematic features can help prevent moment-arm type rotation or pivoting due to applied forces (including gravity).

[0118] In an alternative design where the spatial frame (900) supports the camera channel from a mounting point near the outer polygonal edge of the outer compressor lens element 925, the width of the material between the outer edge of the aperture 915 and the adjacent edge 917 may affect the width of the seam 902 between adjacent camera channels 920. In such cases, where the spatial frame is provided in close proximity to the outer compressor lens element or compressor lens group, optical and mechanical design changes may be advantageous to reduce the impact on the seam width. For example, the outer compressor lens element 925 and lens housing 905 may be designed to conceal the spatial frame 900 so that it does not protrude from the seam into the external environment. Optically, this may mean that the outer compressor lens element 925 bends incident light away from the polygonal lens edge at a very steep angle, thus creating more space relative to the underlying spatial frame. Another option to provide clearance for the spatial frame is a good design that limits the front collar (Figure 5E) to ≤0.5mm, and especially ≤0.1mm.

[0119] Furthermore, as suggested in Figure 2B (see edge 132), the cut or multiple cuts of the polygonal tapered edge 927, or the design of the lens housing 905 and associated lens retaining plate 907, can be modified to provide more room for the spatial frame (900). Additionally, or alternatively, the aforementioned inter-edge width of the spatial frame surface (from the opening 915 to edge 917) can be thinned to reduce its impact on the width of the seam 902 between adjacent camera channels. The design of surface 910 can be less flat and more three-dimensional to provide useful features and improve structural integrity.

[0120] Alternatively, the camera channel can be designed to have an additional enlarged FOV 215 (Figure 7) to optically conceal the extended seam width (902). Depending on the amount of the enlarged FOV 215 added to conceal the extended seam 902 and spatial frame 900, optical correction of residual parallax or perspective errors in the lens design may be compromised. Preferably, these errors remain partial pixels, but for some applications, errors of a few pixels are acceptable as a trade-off for having a spatial frame positioned closer to the periphery of the improved multi-camera panoramic capture device (300).

[0121] Pre-assembled camera channels can be inspected against alignment mounts to help ensure control over size, shape, and datum features. When the improved multi-camera panoramic capture device 300 is then assembled, individual camera channels can be held to or against the relevant faces of the polygonal spatial frame using one or more bolts, magnets, latches, or adhesives, or a combination thereof. However, in some cases, such as a faulty sensor, damaged lens element, or similar, easy replacement of components or camera channels may be necessary. While such replacements can be performed at the factory, field replaceability is desirable. The spatial frame approach can make camera channels field-replaceable units (FRUs).

[0122] One approach to removing one or more camera channels is to allow access to the device through a polygonal facet through which any data and power cabling passes around the spatial frame. If necessary, this cabling can be removed to access the mounting features. A suitable tool can be inserted into the internal hollow central cavity of the spatial frame and extended to the mounting point of the camera channel that needs to be removed, whether for replacement or repair, or to assist in the removal of another camera channel. Given the polygonal shape, the tool used for this purpose may have a flexible shaft or extension to assist in accessing the mounting features or mechanism of the imaging channel, which is angled away from the access port facet. Once released, the camera channel can then be pulled out or removed by pulling it away from the center of the device. The data path and power cabling can then be removed if they have not yet appeared. A pre-assembled replacement camera channel can then be fitted or installed in place of the removed one. If one or more materials or adhesives have been used to seal the joints or gaps between adjacent camera channels, they are also likely to need to be replaced during this process.

[0123] In some cases, the support struts or stalks are first removed from the spatial frame surface, which is the mounting joint. This allows access to the base plate and fasteners within the bottom of the primary assembly. Once the struts are removed, the fasteners or bolts for the primary channel may become accessible. Subsequently, the fasteners for the tertiary channel may be the next most accessible. Finally, the screws for the secondary assembly become accessible. The spatial frame does not need to be disassembled, and the lens housing assembly has kinematic balls, so replacement with a new camera channel assembly should be possible with tight tolerances. In some cases, the camera channel, including the lens housing and lens assembly, is nominally identical. Therefore, only one part number is needed for assembly, storage, and / or replacement.

[0124] When replacing the image sensor assembly in a camera channel, it should be noted that the corresponding circuit board may also need to be replaced, as they are typically a matched set. Furthermore, if the device has a shape other than a dodecahedron and has more camera channels, the size and number of spatial frame planes and camera channels can complicate assembly or repair activities. Specifically, the available space for accessing components and the number of components can be complex. Alternatively, or additionally, for some device designs, a retaining or locking mechanism, such as a latch, can be placed within the seams between imaging channels. In this case, a tool can be inserted into the seam to release the retaining mechanism.

[0125] In other cases, one face of the spatial frame may be damaged and need to be replaced. It can be expected that the associated camera channel, whether damaged or not, will be the first to be removed. However, it may be necessary to remove multiple camera channels first. The retaining mechanism in the attached kinematic element, whether spring, magnet, adhesive, or other mechanism, will then be loosened to release the face and allow the replacement to be installed.

[0126] All the spatial frame examples shown represent dodecahedrons, but other overall shapes are also conceived. For example, spatial frames include, but are not limited to, other polyhedron shapes (including geodesic and Goldberg polyhedra) and can be constructed to form other three-dimensional shapes. Furthermore, while the examples generally deal with structures that provide a 360-degree field of view, in other systems, spatial frames may support different configurations. For example, without limitation, spatial frames can support multiple camera channels of hemispheres, quarter spheres, rings, or other shapes.

[0127] In the previous example, the lens housing and / or spatial frame components (e.g., polygonal faces) can be formed using machined metal, or cast and machined metal such as stainless steel. In other examples, other materials may be used. Some of these alternative materials include plastics, including specially designed materials or composite materials. For example, fiber-reinforced plastics (FRP) are composite materials made of a polymer matrix reinforced with fibers. The fibers are typically glass (in glass fibers), carbon (in carbon fiber reinforced polymers), aramid, or basalt. The polymer is typically epoxy, vinyl ester, or polyester thermosetting plastic. Composite plastics refer to a type of plastic resulting from combining two or more homogeneous materials with different material properties to derive a final product with certain desired material and mechanical properties. Fiber-reinforced plastics are a category of composite plastics that specifically use fibrous materials to mechanically enhance the strength and elasticity of the plastic. Glass fibers are extruded and drawn from materials containing SiO2, Al2O3, B2O3, CaO, or MgO, and then formed into a mat that is subsequently embedded in the polymer during molding, or woven into it. Specifying the orientation of the reinforcing fibers can increase the strength and resistance of the polymer to deformation. Glass-reinforced polymers are strongest and most resistant to deforming forces when the polymer fibers are parallel to the applied force, and weakest when the fibers are perpendicular. In some cast resin components made of glass-reinforced polymers, the fiber orientation can be oriented into two-dimensional and three-dimensional weaves. This means that if the force is probably perpendicular in one direction, they are parallel in another direction, which eliminates the possibility of a weakness in the polymer. Fiber-reinforced plastics are best suited to any design program that requires lightweight, precision technology, constant tolerances, and simplification of parts in both manufacturing and operation. Molded polymer products are cheaper, faster, and easier to manufacture than cast aluminum or steel products, and can be machined after molding while maintaining similar, and sometimes better, tolerances and material strength.

[0128] In some cases, lens housings or spatial frame facets can be molded from fiber-reinforced plastic. The CTE of glass fiber-reinforced plastic varies with the fiber direction, parallel or perpendicular (e.g., 15-55 × 10⁻¹). -6 The CTE (temperature per degree Celsius) is generally in the range between that of glass and that of optical plastics. Lens housings formed from interwoven fibers conveniently have a CTE that is the average between that of optical glass and plastic, thereby supporting the thermal stability of the lens elements located within.

[0129] Further examples may include the manufacture of die-cast alloy lens housings or spatial frame surfaces using zinc-based alloys. Zinc die-casting offers thin walls and excellent surface smoothness. Conveniently, zinc alloys have a CTE (Cold Tensile Energy) relative to that of most optical glass and most optical plastics, thus easily enabling thermal stability of imaging lens assemblies. Die-cast zinc alloy parts can be machined after casting, but typically little machining is required for precision zinc die-castings, as precision can be achieved that is about five times better than that obtained with molded fiberglass parts. In other examples, the material may vary from part to part. For example, some surfaces of a spatial frame may consist of one type of material, while other surfaces consist of a second material.

[0130] In the examples of this disclosure, the polygonal spatial frame is developed as a kinematic structure specifically designed to provide both precise and robust mounting support for an array of camera channels, enabling an improved multi-camera panoramic capture device (300). The spatial frame provides a useful hollow central volume that can be conveniently used by imaging hardware (e.g., power and data path cables) or as an aperture (pass-through) for relayed image light. Alternatively, the supported camera 320 can function as an objective lens providing light to a fiber optic relay rather than a lens system-based imaging relay (e.g., Figure 9). In such a system, the fiber optic relay can use a bundle of fiber optics that transmits image light from each imaging plane provided by the objective lens to the sensor. With respect to image transmission, a coherent fiber optic bundle can be used to relay image light to the image sensor without free-space optical communication. Each camera channel may have an associated image sensor, or several coherent fiber optic bundles may be directed to a single image sensor.

[0131] This type of specially designed kinematic spatial frame may have other optical applications or configurations. For example, it can be used to support an array of cameras that look at an internal, rather than external, hollow center. It can also support other optical devices, such as light sources, lasers, or sensors, which are directed either internally or externally. As another example, the kinematic spatial frame can support an array of light projection channels instead of camera channels. As a feasibility, the image sensor can be replaced with an addressable light source array device having luminescent pixels. For example, these devices could be micro-LED arrays, organic LED arrays (OLEDs), or laser arrays. Alternatively, the image sensor can be replaced with an optical modulator array (e.g., LCOS or DLP / DMD), and the illumination light can be provided by one or more separate light sources. In particular, these spatial frames can be used to enable improved multi-lens systems where the optical design of the lens system is not optimized for parallax or perspective correction, whether for image acquisition, image projection, or other optical purposes. This type of projection device can be used in simulators, planetariums, or other dome theaters. Similarly, in the example, the spatial frame to which the lens is mounted may include a concentric dome or a faceted dome internally. Without limitation, the dome may provide optical properties and casing or covering for the lens element mounted in the frame.

[0132] This type of specially designed kinematic spatial frame can have other non-optical applications or configurations. For example, polygonal spatial frames are also used for interactive toys for children. However, spatial frames are more commonly used in architectural engineering to construct lattice roof supports or geodesic domes. Typically, a geodesic dome is a hemispherical or spherical thin-shell structure (lattice shell) with a polyhedral shape based on a geodesic polyhedron. Engineering emphasis is often directed towards a nodal configuration that attaches to truss joints. The dome can have a steel framework with staves having flat ends, with a single bolt securing the apex of the staves. The dome can also be constructed with a lightweight aluminum framework that can be bolted together, welded, or connected by more flexible nodal / hub connections. The triangular elements of the dome are structurally rigid, distributing structural stresses throughout the structure, allowing the geodesic dome to withstand very heavy loads for its size.

[0133] Geodesic domes can have systems of cables that help maintain the structure together. For this reason, they are called tensile integrity or tensegrity structures, and they consist of elements under simple tension or compression. A subset of these structures may be described as precisely constrained structures, or appear to be, but in reality they may not be statically determinate. Precisely constrained structures, on the other hand, are always statically determinate. Additionally, tensegrity structures may not necessarily be precision structures capable of maintaining tight tolerances, such as those required for most optical applications.

[0134] The specially designed kinematic spatial frame of the present invention may have non-optical applications, including architectural applications. For example, while many skyscrapers are designed to be earthquake-resistant, including by providing compliant frames or foundations, buildings designed and assembled using the type of kinematic spatial frame of the present invention may have improved resistance to earthquakes, wind, or other directional forces. As an architectural structure, the spatial frame can firmly and precisely hold devices in predetermined positions. Alternatively, this approach to specially designed kinematic spatial frames may enable robust, easily assembled emergency shelters (similar to, for example, structures by Shigeru Ban, but different). A roof or covering can be provided over the exterior of the spatial frame.

Claims

1. An imaging device, It is a frame, A first polygonal surface having multiple first edges defining a first periphery, A second polygonal surface having multiple second edges defining a second periphery, One or more kinematic elements connecting a first edge of the plurality of first edges to a first edge of the plurality of second edges, Frames, including A first camera having a first lens connected to the first polygonal surface, wherein the first lens has a plurality of first lens sides defining the periphery of the first lens of the polygon, A second camera connected to the second polygonal surface and having a second lens, wherein the second lens has a plurality of second lens sides defining the periphery of the second lens of the polygon, and the first lens side of the plurality of first lens sides is adjacent to the second lens side of the plurality of second lens sides at a seam, It is an imaging device equipped with, An imaging device in which the one or more kinematic elements include a first kinematic element and a second kinematic element, the first kinematic element connects the second polygonal surface to the first polygonal surface with respect to three degrees of freedom, and the second kinematic element fixes the second polygonal surface to the first polygonal surface with respect to two degrees of freedom.

2. The aforementioned frame is A third polygonal surface having multiple third edges defining a third periphery, An additional kinematic element is provided, which connects the first edge of the plurality of third edges to the second edge of the plurality of first edges, and connects the second edge of the plurality of third edges to the second edge of the plurality of second edges. It further includes, The imaging device is A third camera connected to the third polygonal surface. The imaging apparatus according to claim 1, further comprising:

3. The aforementioned frame is To form a kinematic spatial frame, it further includes additional polygonal faces connected to at least one of the first polygonal face, the second polygonal face, and the third polygonal face, The imaging device is An additional camera connected to the additional polygonal surface, wherein the first camera, the second camera, the third camera, and the additional camera are configured to provide substantially adjacent fields of view. The imaging apparatus according to claim 2, further comprising:

4. The one or more kinematic elements are, A cylinder pin having a vertical axis extending substantially parallel to the first edge of the plurality of first edges and the first edge of the plurality of second edges, wherein the outer surface of the cylinder pin is in contact with the first datum surface on the first edge and the second datum surface on the second edge, The first polygonal surface and the first spring connected to the cylinder pin, The second polygonal surface and the second spring connected to the cylinder pin, The imaging apparatus according to claim 1, including the following:

5. The imaging apparatus according to claim 1, wherein the first periphery is the same as the periphery of the first lens, and the second periphery is the same as the periphery of the second lens.

6. The imaging apparatus according to claim 1, wherein one of the one or more kinematic elements includes a first portion fixed to one of the first polygonal surface or the second polygonal surface and a second portion fixed to the other of the first polygonal surface or the second polygonal surface, the first portion includes a ball, and the second portion includes at least one of a flat portion, a V-shaped groove, and a three-sided socket that is contacted by the ball.

7. The imaging apparatus according to claim 1, wherein the one or more kinematic elements include a cylinder pin and an offset V-shape.

8. The imaging apparatus according to claim 1, further comprising a magnet or spring associated with one or more kinematic elements to provide holding force.

9. The imaging device according to claim 1, further comprising a ball disposed along a first edge around the first lens and at least one of a flat portion and a V-shape disposed along a second edge around the second lens, wherein the ball is in contact with at least one of the flat portion and the V-shape to maintain the seam between the first lens and the second lens.

10. The imaging apparatus according to claim 1, wherein at least one of the first cameras is coupled to the first polygonal surface, or the second camera is coupled to the second polygonal surface, using at least one of a magnet, a V-shaped groove, a flat portion, and an alignment ball.

11. The imaging apparatus according to claim 1, wherein the first camera is positioned relative to the second camera such that a first low parallax amount associated with the first camera at least partially overlaps with a second low parallax amount associated with the second camera.

12. The imaging device according to claim 1, wherein the frame defines the central aerial center.

13. A relay optical system that extends at least partially into the hollow center and passes through an opening in the plane of the frame facing the first polygonal surface. The imaging apparatus according to claim 12, further comprising:

14. A frame for an imaging device, A first polygonal surface having multiple first edges defining a first periphery, A second polygonal surface having multiple second edges defining a second periphery, One or more kinematic elements connecting a first edge of the plurality of first edges to a first edge of the plurality of second edges, It is a frame equipped with, A frame comprising one or more kinematic elements, including a first kinematic element and a second kinematic element, wherein the first kinematic element connects the second polygonal surface to the first polygonal surface with respect to three degrees of freedom, and the second kinematic element fixes the second polygonal surface to the first polygonal surface with respect to two degrees of freedom.

15. One of the kinematic elements among the one or more kinematic elements is A cylinder pin having a vertical axis extending substantially parallel to the first edge and the second edge, wherein the outer surface of the cylinder pin contacts the first datum surface on the first edge of the plurality of first edges and the second datum surface on the first edge of the plurality of second edges, The first polygonal surface and the first spring connected to the cylinder pin, The second polygonal surface and the second spring connected to the cylinder pin, The frame according to claim 14, including the frame described in claim 14.

16. A third polygonal surface having multiple third edges defining a third periphery, An additional kinematic element that connects the first edge of the plurality of third edges to the second edge of the plurality of first edges, and connects the second edge of the plurality of third edges to the second edge of the plurality of second edges, The frame according to claim 14, further comprising:

17. An additional polygonal surface connected to at least one of the first polygonal surface, the second polygonal surface, and the third polygonal surface in order to form a rigid kinematic spatial frame. The frame according to claim 16, further comprising:

18. The frame according to claim 14, further comprising one or more magnets or springs associated with the one or more kinematic elements in order to provide a holding force.

Images