Patent Application
20110242272
The design and fabrication of imaging systems or “imagers” (e.g., cameras) incorporating image detector arrays such as charge-coupled devices (CCDs) and complimentary metal-oxide semiconductor (CMOS) circuits is an established art. Such detector arrays are typically formed on a planar substrate and are, therefore, frequently referred to as “focal plane arrays.” Depending on the particular application for which an imager is designed, different optical components are implemented to optically communicate images to the image detector array. For example, in a simple consumer digital video or still-image camera, at least one lens is situated forward of the focal plane array for projecting images of scenes being filmed onto the image detector array. In such a case, the at least one lens defines an optical axis that is oriented orthogonally to the plane defined by the focal plane array.
In creating imagers for more complex applications, such as video surveillance, imaging-system designers are confronted with the task of communicating “wide-angle” or “panoramic” images to a single focal plane array. Beyond certain limits, the use of a single wide-angle lens results in unacceptable image distortion. In recognition of the image distortion introduced by large-field-of-view lenses, attempts have been made to assemble imaging systems with multiple, disparately-directed lenses, the individual images from which are then focused onto one or more detector arrays, digitized and “combined” by techniques such as algorithmic “correction” and pixel matching. The goal of such devices is to create a corrected image that is a digitized representation of a continuous region of space representing a large field of view with reduced distortion.
The use of multiple disparately-directed lenses combined with algorithmic correction presents its own set of challenges. According to one approach, undistorted images can be captured and stored in computer memory if multiple focal plane arrays are used. In such a case, each lens is optically aligned with its own focal plane array. However, such an implementation introduces the complexities of communicating to computer memory, synchronizing and algorithmically analyzing the signal outputs of multiple focal plane arrays. Moreover, electronically packaging multiple disparately angled focal plane arrays invites a host of challenges, not the least of which is spatial efficiency. In a second approach, multiple disparately oriented lenses are used to simultaneously project a corresponding number of images onto a single focal plane array. A particular implementation is represented by the so-called “bug eye” by BAE Systems. In the “bug eye” system, several—specifically nine (9)—lenses are arranged on a hemispherical surface. A faceted fused optical fiber bundle is interposed between the lenses and the focal plane array in order to compensate for extreme non-perpendicularity between the focal plane array and peripheral lenses. One image is projected from each of the nine lenses onto a corresponding facet of the fused bundle and then conveyed by internal reflection to a corresponding image-detecting portion of the focal plane array. While the “bug eye” obviates the issues associated with the use of multiple disparately angled focal plane arrays, only one of the nine lenses defines an optical axis that is oriented orthogonally to the focal plane array and/or a central facet of the faceted bundle. Each of the other eight lenses defines an optical axis that is oriented at an angle of substantially less than 90° relative to the focal plane array and to the facet with which that lens is aligned. Consequently, although the image axis defined by each of the “non-orthogonal” lenses is within the acceptance angle of the fiber ends defining a corresponding bundle facet, and none of the nine lenses individually is sufficiently “wide-angled” to introduce appreciable distortion, there is nevertheless substantial image distortion associated with each of the “non-orthogonal” lenses by virtue of the fact that the optical axis associated with each such lens is oriented at a shallow angle with respect to its corresponding bundle facet.
In recognition of the need to contemporaneously deliver multiple undistorted images to a single focal plane array, one prior design employs flexible image guides. More specifically, the “Poly Optical Fiber Device” introduced by Volpi Manufacturing USA (hereinafter, “Volpi”) allows up to eight (8) different views to be conveyed simultaneously to a single camera through flexible coherent fiber-optic imaging guides. In associated product literature, Volpi emphasizes the flexibility of each “fiber leg,” and how that flexibility facilitates the reorientation of legs relative to one another. The flexibility of individual imaging guides is undoubtedly an advantage in some applications. However, if a device such as Volpi's Poly Optical Fiber device is to be employed under conditions requiring fixed imaging angles, then an “exoskeletal” structure or framework is required in order to retain each leg in a fixed position and angular orientation.
Accordingly, there exists a need for a self-supporting multi-directional imaging assembly that contemporaneously communicates multiple minimally-distorted images to a single planar detector array.
A central component of an illustratively embodied multi-directional imaging assembly is a multi-directional imaging bundle. In one embodiment, a multi-directional imaging bundle has a plurality of at least two image-conducting branch elements. Each branch element has an imaging-conducting first portion with an image-input face and an image-conducting second portion with an image-output face opposite the image-input face. Moreover, each branch element is rigid over its entire length between the opposed image-input and image-output faces and includes at least one bend between the first and second portions such that the first and second portions extend along, respectively, a first image-propagation axis and second image-propagation axis that is non-parallel to the first image-propagation axis.
A multi-directional imaging bundle is formed by mutually binding the second portions of at least first and second image-conducting branch elements. In alternative illustrative versions, mechanical binding of the branch elements is accomplished by at least one of (i) heat fusing and (ii) application of an adhesive, such as an epoxy, by way of non-limiting example. The mutually bound second portions define a bundle trunk and extend along, though not necessarily parallel to, a common bundle axis. Furthermore, the image-output faces of the branch elements coincide in order to define a common image-output face. It will be appreciated that the rigidity of the individual branch elements renders the imaging bundle self-supporting, thereby obviating the need for an “exoskeletal” framework for supporting the individual branch elements to keep them aligned as desired.
When the second portions of the branch elements are mutually bound as described above, the first portions of the branch elements are mutually divergent. Additionally, in various versions, the image-input faces are disparately directed. For example, in a non-limiting illustrative instance in which the image-input face of each of the first and second branch elements is planar and oriented orthogonally to the first image-propagation axis of that branch element, it will be readily appreciated that the image-input faces of the first and second branch elements are disparately directed. However, as implied by the non-limiting nature of the aforementioned example, it is to be expressly understood that versions with alternatively configured and directed image-input faces are within the scope and contemplation of the invention. More specifically, by way of additional non-limiting example, within the scope and contemplation of the invention are versions in which a planar image-input face is not oriented orthogonally to a corresponding first image-propagation axis. Additionally, in the absence of express claim language to the contrary, versions with non-planar image-input faces are also regarded as within the scope of the invention as defined in the appended claims.
Aspects of illustrative multi-directional imaging bundles having been described, an illustrative multi-directional imaging assembly further includes an optical focusing element mechanically retained in optical alignment with the image-input face of each branch element. In a typical version, each focusing element is a lens. However, absent explicit limitations to the contrary, it is to be understood that within the scope and contemplation of the invention as defined in the appended claims are versions having alternative focusing optics such as, by way of non-limiting example, mirrors or graded refractive index elements. Each focusing element defines a field of view correlating to a three-dimensional region of space external to the imaging assembly. An image of the spatial region correlating to the field of view defined by a focusing element is acquired and projected by that focusing element onto the image-input face with which that focusing element is optically aligned.
The images acquired and projected by each focusing element are conducted by internal refection through the branch element with which that focusing element corresponds. Accordingly, in various versions, each image-conducting branch element comprises a plurality of adjacently fused, internally-reflecting imaging conduits, such as optical fibers. The heating, drawing and adjacent fusing of bundled optical fibers, or optical fiber canes, to form rigid, image-conducting bundles is a well-established art and, therefore, warrants no detailed explanation herein. A straight bundled formed by such a process would be bent before it cools or, if cooled, subsequently heated and bent to form an image-conducting branch element including at least one bend.
The field of view defined by each focusing element is unique relative to the field of view defined by each of the other focusing elements associated with the imaging assembly. However, in order to render possible the “piecing together” of plural images acquired by plural focusing elements, and create a composite image representing a single, continuous region of space, each of various embodiments is configured such that the field of view defined by each focusing element partially overlaps the field of view defined by at least one other focusing element associated with the imaging assembly. For example, a left-side focusing element might include within the right side of its field of view some of the same portions of space included within the left side of the field of view of a right-side focusing element. The left and right side images could then be combined, and the redundancy between them “subtracted out” out of a combined image by, for example, algorithmically-based image enhancement techniques. The goal of the image enhancement is to create, in a “combined” or “compound” image, the illusion of a single panoramic image acquired over a larger field of view than any of the focusing elements is individually capable of yielding.
Situated in optical alignment with the common image-output face of the imaging bundle is an image detector array to which images conducted through the branch elements and outputted through the common image-output face are optically communicated. The image detector array used in any particular version may be a (i) microbolometer, (ii) a charge-coupled device (CCD), a (iii) complimentary metal-oxide semiconductor (CMOS) circuit or (iv) one or more photodiodes, by way of non-limiting example. An array of at least 1×1 photodiodes might be used as the detector array in an embodiment configured for general light detection. Accordingly, in at least one such embodiment, focusing elements are omitted at the image-input faces since “focused” images might be unnecessary for such purposes. In any event, in an illustrative version in which a detector array of any type is included, the detector array is communicatively linked to a data processing system including a central processor, memory for storing data indicative of images registered by the detector array (alternatively, “registered-image data”) and a signal processing algorithm for processing the electrical outputs of the detector array.
Registered-image data representative of multiple (at least two) images registered simultaneously at the detector array is stored in computer memory. It will be appreciated that, by virtue of the overlapping fields of view aforementioned, that some of the registered-image data associated with at least one image will be duplicative of some of the registered-image data associated with at least one other image. Accordingly, in at least some implementations, a signal-processing algorithm analyzes the registered-image data corresponding to contemporaneously, acquired images and algorithmically assembles a single composite image in which image-data redundancy is eliminated.
Representative, non-limiting embodiments are more completely described and depicted in the following detailed description and the accompanying drawings.
The following description of various rigid multi-directional imaging bundles and variously embodied multi-directional imaging assemblies incorporating multi-directional imaging bundles is illustrative in nature and is therefore not intended to limit the invention or its application of uses. The various implementations, aspects, versions and embodiments described in the summary and detailed description are in the nature of non-limiting examples falling within the scope of the appended claims and do not serve to constrain the maximum scope of the claims.
Referring to
Images outputted through the common image-output face 105 are optically communicated to an image detector array 110. In the version depicted, the common image-output face 105 and image detector array 110 are mutually coupled “directly” such that no intervening optics are required. However, it is to be understood, absent express limitations to the contrary, versions in which at least one optical element (not shown) is present between the image-output face 105 and the image detector array 110 are within the scope and contemplation of the invention as defined in the appended claims.
Alternative implementations incorporate any of a variety of conventional detector arrays 110 configured to detect wavelengths over a predetermined range of electromagnetic wavelengths. A typical detector array 110 suitable for implementing embodiments of the invention includes photosensitive detector elements 115 that are, to the extent practicable, uniformly sized and regularly spaced. As mentioned in the summary, three illustrative types of detector arrays 110 that may be incorporated into various alternative embodiments are (i) microbolometers (ii) charge-coupled devices (CCD) and (iii) complimentary metal-oxide semiconductor (CMOS) circuits. The detector array 110 is communicatively linked to a data processing system 200 including a central processor 205, memory 210 for storing data indicative of registered images 215 (alternatively, “registered-image data 215”), and a signal processing algorithm 220 for processing the electrical outputs of the detector array 110 and the registered-image data 215.
In the illustrative version of
Referring still to
With the second portions 140 of the branch elements 120 mutually bound, the first portions 130 of the branch elements 120 are mutually divergent relative to the second portions 140 and the bundle trunk 106 defined thereby. More specifically, while the second portions 140 might not, in any particular version, be parallel to one another, and to the common bundle axis AB, the second portions 140 of the first and second branch elements 120A and 120B are closer to mutually parallel than are the first portions 130 of the first and second branch elements 120A and 120B. Additionally, in various versions, the image-input faces 132 of the first and second branch elements 120A and 120B are disparately directed. For instance, in the particular, non-limiting version of
Mechanically retained in optical alignment with the image-input faces 132 of the first and second branch elements 120A and 120B are, respectively, first and second focusing elements 160A and 160B. As with the first and second branch elements 120A and 120B, when one focusing element 160 need be distinguished from the other, the alphabetic characters “A” and “B” included. As schematically depicted in
At a given instant, a first image I1 of the first spatial region SR1 is acquired and projected by the first focusing element 160A onto the image-input face 132 of the first branch element 120A. Contemporaneously with the acquisition and projection of the first image I1, a second image I2 of the second spatial region SR2 is acquired and projected by the second focusing element 160B onto the image-input face 132 of the second branch element 120B. Subsequent to their impingement upon the image-input faces 132 of the first and second branch elements 120A and 120B, the first and second images I1 and I2 are conducted by internal reflection through, respectively, the first and second branch elements 120A and 120B and optically communicated to the detector elements 115 of the detector array 110 through the common image-output face 105.
Although the internally-reflecting branch elements 120 may be alternatively configured, each of the branch elements 120 of
By virtue of the disparate directing of the focusing optics, the field of view defined by the first focusing element 160A differs from the field of view defined by the second focusing element 160B. That is to say, the first and second focusing elements 160A and 160B acquire and project images of disparate first and second spatial regions SR1 and SR2 such that a first image I1 acquired and projected by the first focusing element 160A differs from a second image I2 simultaneously acquired and projected by the second focusing element 160B. However, the first and second focusing elements 160A and 160B are configured and directed such that the first field of view FOV1 partially overlaps the second field of view FOV2. This overlap is represented in
Registered-image data 215 representative of first and second images I1 and I2 registered simultaneously at the detector array 110 is stored in computer memory 210. It will be appreciated that, because some of the “object content” of the first image I1 is the same as some of the object content of the second image I2 acquired contemporaneously, there will exist some redundancy in the registered-image data 215 indicative of the first and second image I1 and I2. Accordingly, in at least some implementations, a signal-processing algorithm 220 analyzes the registered-image data 215 corresponding to the first and second images I1 and I2 in order to algorithmically construct (or assemble) a composite image IC in which image-data redundancy is eliminated. At the bottom of
Although the illustrative embodiment of
The foregoing is considered to be illustrative of the principles of the invention. Furthermore, since modifications and changes to various aspects and implementations will occur to those skilled in the art without departing from the scope and spirit of the invention, it is to be understood that the foregoing does not limit the invention as expressed in the appended claims to the exact constructions, implementations and versions shown and described.
Priority based on Provisional Application Ser. No. 61/340,732 filed Mar. 22, 2010, and entitled “RIGID, MULTI-DIRECTIONAL IMAGING BUNDLE AND IMAGING ASSEMBLY INCORPORATING THE SAME” is claimed. The entirety of the disclosure of the previous provisional application, including the drawings, is incorporated herein by reference as if set forth fully in the present application.
| Number | Date | Country | |
|---|---|---|---|
| 61340732 | Mar 2010 | US |
