FIELD OF THE INVENTION

In general, the present invention relates to a method and apparatus for selectively reducing in real-time, the intensity of incident light rays propagating towards an optical element such as an eye, a camera station in a machine vision system or other image detection device.

More particularly, the present invention relates to an intelligent electro-optic system which automatically eliminates glare produced by intense sources of illumination present in the environment.

BACKGROUND OF THE INVENTION

It is well known that intense sources of illumination can produce glare which impairs the operation of various types of optical systems.

For example, automobile drivers at night face the hazard of glare produced when intense light from the headlamps of oncoming cars impinges onto their eyes. When the headlamps of oncoming vehicles are operated in their high-beam mode, the light produced therefrom is often so intense as to disable the driver from viewing the road ahead. This problem tends to worsen as the driver's eyes are exposed to prolonged periods of headlamp illumination. Consequently, the driver's vision becomes fatigued and impairing his or her ability to drive effectively and thus presenting hazardous situations.

The above problem is particularly severe for the increasing elderly population. It is well known that older men and women tend to lose their ability to adapt to rapid changes in light intensity, making them more vulnerable to intense illumination produced from oncoming headlamps. Consequently, many older men and women avoid driving at night, depriving them the freedom to do things they once used to do.

OBJECTS OF THE INVENTION

Accordingly, it is a primary object of the present invention to provide an intelligent electro-optic system which automatically reduces the intensity of incident light rays from intense sources of illumination.

A further object of the present invention is to provide an intelligent electro-optical system which automatically modulates the intensity of light rays from propagating from points of illumination in spatial scenery to the eyes of the user viewing the spatial scenery.

A further object of the present invention is to provide an intelligent electro-optical system for reducing glare while operating an automotive vehicle, yet without impairing the driver's vision or ability to drive effectively.

Another object of the present invention is to provide an intelligent electro-optical system which automatically reduces glare when a driver views spatial scenery through either the front windshield, the rear view mirror, or the side view mirrors of an automotive vehicle.

Yet a further object of the present invention is to provide a portable electro-optical system in which a camera station is embodied in a head support frame having a pair of optically transparent electro-optical lenses, each disposed in front of one of the user's eyes to selectively filter out sources of solar and/or headlight glare.

These and further objects of the present invention will become apparent hereinafter.

SUMMARY OF THE INVENTION

In accordance with one of the broader aspects of the present invention, the method and apparatus are provided for selectively reducing the intensity of light rays as they propagate from a spatial scene towards an optical element, such as an eye or camera, having a field of view.

In general, the apparatus comprises an electro-optical element, image acquisition means, image processing means and control means. The electro-optical element has an optically transparent surface consisting of a plurality of pixels. Each pixel has a controllable light transmittance for selectively reducing the intensity of light rays propagating from a point in a spatial scene, through the pixel, then towards the optical element. The image acquisition means is provided for acquiring an image of the spatial scene within the field of view of the optical element. The image processing means is provided for processing the image and determining at which pixels the light transmittance is to be actively controlled in order to reduce the intensity of incident light rays by a selective amount before they reach the optical element. The control means is provided for actively controlling the light transmittance of the determined pixels so that as incident light rays propagate through the determined pixels, the incident light rays impinge the optical element with an intensity reduced by the selected amount.

BRIEF DESCRIPTION OF THE DRAWINGS

For a full explanation of the present invention, the following Detailed Description of the Illustrated Embodiments is to be read in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic diagram illustrating a generalized electro-optical system of the present invention, in which the intensity of an incident light ray propagating from a point of illumination in a 3-D spatial scene towards an optical element, is reduced in intensity at the point of intersection through the light intensity reducing surface of the system prior to reaching the optical element;

FIG. 2 is a schematic diagram illustrating the configuration of a first embodiment of the electro-optical system of the present invention, in which the light modulating surface is an optically transparent liquid crystal light valve panel carrying a stereo scene-imaging subsystem for imaging spatial scenery within the field of view of an automobile driver, and a stereo pupil-tracking camera subsystem for measuring the position of the driver's pupils relative to the liquid crystal light valve panel;

FIG. 3 is a schematic diagram illustrating the operation of the electro-optical system of FIG. 2, as light rays propagate from a point of illumination in the spatial scene, through the liquid crystal light valve panel and then intensity reduced prior to passing through the pupils of the driver's eyes;

FIG. 4A is a schematic representation of an image of a driver's face produced by a camera station in the pupil-tracking camera subsystem;

FIG. 4B is a schematic representation of an enlarged image of the driver's pupil represented in FIG. 4A;

FIG. 4C is a schematic diagram of a camera station employed in the stereo camera subsystems of the electro-optical system of the present invention;

FIG. 5A and 5B is a flow chart showing the steps performed in determining the pixel locations of the liquid crystal light valve panel of the system of FIG. 2, which are electrically addressed and actively controlled in order to reduce the intensity of light rays propagating from a point of illumination in the spatial scene, towards the eyes of the driver;

FIG. 6 is a schematic diagram illustrating the configuration of a second embodiment of the electro-optical system of the present invention, in which the light intensity reducing surface is a reflective-type liquid crystal light valve panel carrying a stereo scene camera subsystem for imaging spatial scenery within the field of view of an automobile driver, and a stereo pupil-tracking camera subsystem for measuring the position of the driver's pupils relative to the liquid crystal light valve panel;

FIG. 7 is a schematic diagram illustrating the operation of the electro-optical system of FIG. 6, as light rays propagate from a point of illumination in the spatial scene, through the liquid crystal light valve and then intensity reduced prior to passing through the pupils of a driver's eyes;

FIG. 8A and 8B is a flow chart showing the steps performed in determining the pixel locations of the liquid crystal light valve panel of the system of FIG. 6, and which are electrically addressed and controlled when reducing the intensity of light rays propagating from points of illumination in the spatial scene, toward the eyes of the driver;

FIG. 9 is a schematic diagram illustrating the configuration of a third embodiment of the electro-optical system of the present invention, in which the light intensity reducing surface is an optically transparent liquid crystal light valve, and the driver carries a stereo scene imaging camera subsystem on his head for imaging for spatial scenery within the driver's field of view;

FIG. 10 is a schematic diagram illustrating the operation of the electro-optical system of FIG. 9 as light rays propagate from a point of illumination in the spatial scene through the liquid crystal light valve panel and then intensity reduced prior to passing through the pupils of a driver's eyes;

FIG. 11A, 11B and 11C is a flow chart showing the steps performed in determining the pixel locations of the liquid crystal light valve panel of the system of FIG. 9, which are electrically addressed and actively controlled in order to reduce the intensity of light rays propagating from points of illumination in the spatial scene, towards the eyes of the driver;

FIG. 12 is a schematic diagram illustrating the configuration of a fourth embodiment of the electro-optical of the present invention, in which the light intensity reducing surface is a reflective-type liquid crystal light valve panel, and the driver carries a stereo scene imaging camera subsystem on his head for imaging spatial scenery within the driver's field of view;

FIG. 13 is a schematic diagram illustrating the operation of the electro-optical system of FIG. 12, as light rays propagating from a point of illumination in the spatial scene reflects off the liquid crystal light valve panel and then intensity reduced prior to passing through the pupils of the driver's eyes;

FIG. 14A, 14B and 14C is a flow chart showing the steps performed in determining in the pixel locations of the liquid crystal light valve panel of the system of FIG. 12, which are electrically addressed and actively controlled pixels in order to reduced the intensity of light rays propagating from points of illumination in the spatial scene, towards the eyes of the driver;

FIG. 15 is a schematic diagram of a fifth embodiment of the electro-optical system of the present invention, in which the light intensity reducing surface is a optically o transparent liquid crystal light valve panel carrying a monocular scene-imaging camera for imaging spatial scenery within the field of view of an automobile driver, and an eye and head tracking subsystem for measuring the position and orientation of the driver's eyes relative to the liquid crystal light valve panel; and

FIG. 15A is a plan view of the electro-optical system illustrated in FIG. 15.

DETAILED DESCRIPTION OF THE ILLUSTRATIVE EMBODIMENTS

In FIG. 1, the apparatus of the present invention is generally illustrated in the form of an intelligent electro-optical system. The primary function of electro-optical system is to selectively reduced (i.e. decrease) the intensity of incident light rays as they propagate from a three-dimensional (3-D) spatial scene towards an optical element 1. In accordance with the invention, the optical element may be the pupil of a human eye, or the aperture stop or lens of a camera station, machine vision system or any other image detection device. As illustrated, the optical element is disposed at a selected distance from the spatial scene and has a field of view in the direction thereof.

In general, the electro-optical system of the present invention comprises a number of system components, namely: an electro-optical element 3, an image acquisition means 4, an image processing means 5, and a control means 6 operably associated as shown. As illustrated, the electro-optical element has an optically transparent surface consisting of a large number of optically transparent pixels 5. Each of these pixels is electrically addressable by controller 6, and has a light transmittance which is actively controllable for the purpose of selectively reducing the intensity of an incident light ray r_j propagating from a point P_i in the spatial scene, through the pixel, towards the optical element. Typically, points on the surface of electro-optical element 3 are measured with respect to a coordinate system R_s. While the coordinate system R_s is a Cartesian coordinate system specified by principal coordinate axes x, y and z, it is understood that other types of coordinates systems may be used.

Preferably, the pixels along the transparent surface are formed from a polymer-dispersed liquid crystal film having a light transmittance of at least 70% in the optical spectrum, that is, when the pixels are not actively controlled or driven by controller 6. Each pixel located on the optically transparent surface at coordinates (x, y) is electrically addressable by an address value A(x,y) computed by computer system 5. When driving an addressed pixel or set of pixels at any particular instant in time, the intensity (i.e. amplitude) of incident light rays falling upon these actively driven pixels is intensity reduced by a selected amount which is sufficient to achieve effective reduction of glare produced in diverse environments. The degree of intensity reduction achievable at each pixel can be of a binary nature (i.e., a first light transmittance when not actively driven, or a lesser light transmittance when actively driven). Alternatively, the degree of intensity reduction m(x, y) can be quantized to one of a number of possible states. For further details regarding suitable polymer-dispersed liquid crystal films that may be used in practicing the present invention, reference is made to the following publications: "Reverse-Mode MicroDroplet Liquid Crystal Display" by Y. D. Ma and B. G. Wu, on pages 46-57, SPIE Vol. 1257, Liquid Crystal Displays and Application (1990); and "Polymer-Dispersed and Encapsulated Liquid Crystal Films", by G. Paul Montgomery, Jr., on pages 577-606, SPIE Institute Series Vol. IS4, Large-Area Chromogenics: Materials and Devices for Transmittance Control 1990, which are hereby incorporated by reference.

In general, image acquisition means 4 is realizable as a camera station having image forming optics and a CCD image detection array. The coordinates of pixels on the image detection array are measured with respect to coordinate system R_c. As illustrated, a coordinate system R_oe, specified by principal axes x', y' and z', is embedded in the optical element for measuring the position of points therein. The principal function of the camera station is to acquire images of the spatial scene within the field of view of the optical element. While not essential to the invention, these images may be acquired along the field of view of the optical element when as viewed, for example, through the optically transparent surface of electro-optical element 3, as shown in FIG. 1.

Image processing means 5 is realizable as a microcomputer system having associated memory for buffering acquired images. The microcomputer processes the acquired image(s) from the camera station in order to determine at which pixels in the electro-optical surface, the light transmittance is to be actively controlled in order to reduce the intensity of incident light rays by a selected amount before they reach the optical element. The microcomputer produces intensity reduction data m(x,y) representative of the selected amount of intensity reduction at the pixels located at coordinates x, y.

Control means 6 is realizable as controller/driver circuitry interfaced with the microcomputer 5. The principal function of control means 6 is to automatically address particular pixels and actively control the light transmittance thereof accordance with intensity reduction data m(x,y). In this way, as light rays propagate from the spatial scene and through the actively controlled pixels in electro-optical surface 3, the incident light rays propagating through these pixels will reach the optical element with an intensity that has been reduced by the selected amount of light transmittance (e.g. 30%).

Referring to FIGS. 2 through 5B, the first illustrated embodiment of the electro-optical system of the present invention will be described. In this and other embodiments illustrated hereinafter, the optical element(s) being "protected" by the electro-optical system of the invention are the eyes of an automobile driver. It is understood however, that the optical element(s) may be the camera station(s) of a machine vision system, or any image detection device desiring or requiring protection from intense sources of environmental illumination.

As illustrated in FIG. 2, the electro-optical element of system 7 is in the form of an optically transparent liquid crystal light valve (LCLV) panel 8 adapted for pivotal mounting inside an automobile above the dashboard in a manner similar to a conventional sunvisor. Preferably, the width of LCLV panel 8 is about 60 centimeters, the height thereof about 10 centimeters, and the size of each pixel about 2.5 millimeters, although it is well understood these dimensions will vary from embodiment to embodiment. On opposite sides of the upper portion of the LCLV panel, a pair of infrared cameras 9A and 9B are mounted with their optical axes directed towards the automobile driver in order to form a pupil-tracking camera subsystem which measures the position of the driver's pupils 10A and 10B relative to the LCLV panel. In general, each camera station comprises image forming optics and a CCD image detection array responsive to the infra-red radiation naturally emitted from the driver's eyes. Techniques for determining the position of the driver's pupils from an acquired pair of stereo infrared images, are well known in the art and are described in detail in Chapter 13, "Photogrammetry & Stereo", on pages 299-327 of Robot Vision (1991) by Berthold Klaus Paul Horn, published by MIT Press, Cambridge, Massachusetts.

As illustrated in FIGS. 4A and 4B, the method of pupil position determination generally involves first recognizing the driver's eye 11 and then the center portion 11A of the pupil. Then pixel data at center 11A is processed by computer 12 using known stereo image processing techniques, to produce coordinate data x, y, z corresponding to the position of the pupil. This process is performed for both eyes to continuously provide coordinate data regarding the position of the driver's pupils.

On opposite sides the upper portion of the LCLV panel, adjacent infra-red cameras 9A and 9B, a pair of camera stations 13A and 13B are mounted with their optical axes directed away from the automobile driver, into the direction of oncoming traffic. Together, these camera stations form a stereo scene-imaging camera subsystem which images spatial scenery within the driver's field of view extending through the automobile windshield. As with the infra-red camera stations, Camera stations 13A and 13B include image forming optics 14 and a CCD image detection array 15, such as shown for example in FIG. 4C. On the basis of a stereo image pair acquired by the scene-imaging camera subsystem, depth-map data (i.e. x, y and z coordinates) of each point in the spatial scene can be readily computed by computer 12 using techniques discussed in detail in Chapter 13 of Robot Vision, supra. Upon processing the captured stereo images of the spatial scene and the computed pupil position data, computer 12 generates addresses A(x,y) and intensity reduction data m(x,y) corresponding to those pixels which require a change in light transmittance. This data is provided to controller/driver 16 which, in turn, addresses particular pixels coinciding with high intensity points in the spatial scene. The controller/driver then drives these addressed pixels in order to actively control the light transmittance thereof and decrease the intensity of incident light rays while propagating through actively driven pixels. As a result, the spatial scenery viewed by the automobile driver is automatically devoid of glare commonly associated with intense sources of illumination, such as the headlamps of oncoming vehicles.

In FIG. 3, a ray optics model is presented for the electro-optical system of FIG. 2. As shown, coordinate system R_c, specified by x, y and z axes, is embedded within the LCLV panel so that (i) the x-y plane lies within the planar surface of the LCLV panel, (ii) the x axis is perpendicular to the optical axes of camera stations 9A, 9B, 13A and 13B, and (iii) the z axis is parallel with the optical axes of camera stations 13A and 13B and antiparallel with camera stations 9A and 9B. Typically, the position of pixels in the CCD array of the pupil-tracking and scene-imaging camera substations are specified by coordinates measured in local coordinate systems (not shown), and can be converted to coordinate system R_c using homogenous transformations well known in the art.

Disposed at the center of the pupil of the driver's left is the origin of coordinate system R_LE, which is specified by axes x', y' and z', with the z' axis aligned with the optical axis of the driver's left eye. Disposed at the center of the pupil of the driver's right eye is the origin of coordinate system R_RE, which is specified by axes x", y" and z" with the z" axis aligned with the optical axis of the driver's right eye and the y" axis aligned with the x' axis of coordinate system R_LE, as shown. The position of origin O_LE of coordinate system R_LE, is specified by position vector t_L defined in coordinate system R_c. Similarly, the position of the origin O_Re of coordinate system R_RE is specified by position vector t_r defined in coordinate system R_c. Points P_i in the scene are specified in coordinate system R_c by coordinates (x_i, y_i, z_i), and points in the scene having an intensity above a predetermined threshold are designed as P_j. The position of point P_i in coordinate system R_c is specified by coordinates (x_j, y_j, z_j).

A light ray propagating from a point P_j in the spatial scene (by a reflection or radiation process), through the transparent surface of LCLV panel 8 and then finally towards the pupil of the driver's left eye, is specified by a 3-D line equation, r_Lj. Similarly, a light ray propagating from a point P_j in the spatial scene, through the LCLV panel, and then finally towards the pupil of the driver's right eye, is specified by a 3-D line equation, r_rj. The point at which light ray r_Lj intersects the optically transparent surface of the LCLV panel is designated by Q_Lj, whereas the point at which light ray r_rj intersects the transparent surface is designated by Q_rj.

As the driver views spatial scenery through both the LCLV panel and the automobile window, a number of steps are repeatedly performed by the intelligent electro-optical system 7, permitting the driver to operate the automobile, while light reducing (e.g. blocking) pixel locations are automatically determined, electrically addressed and actively driven on a real-time basis to reduce glare associated with oncoming headlights.

In FIGS. 5A and 5B, a generalized process is described for automatically modulating the intensity of light rays propagating through the LCLV panel of electro-optical system 7, towards the eyes of the automobile driver. As indicated at block A of FIG. 5A, the first step of the process involves forming a 3-D depth map of the spatial scene being viewed by the driver through the LCLV panel. This step results in determining the x_i, y_i, z_i coordinates for each point P_i in the spatial scene, measured with respect to coordinate system R_c. As indicated at block B, computer 12 determines which points P_i in the imaged scene have a detected intensity above a preselected intensity threshold, and assigns to each of these points P_j, a desired intensity reduction m(P_i). Then, as indicated in block C, the computer processes acquired images of the driver's eyes in order to determine the position coordinate of the pupil centers, i.e. O_Le, O_re, measured with respect to coordinate system R_c.

As indicated at blocks D through H, the computer uses the coordinates of point P_j (derived from the pair of scene images) and the pupil position coordinates (derived from the pair of eye images) in order to formulate a representation of a light ray path extending from point P_j in the spatial scene to the driver's left pupil. The computer then computes the pixel coordinates through which the light ray r_Lj intersects, and then LCLV controller/driver 16 electrically addresses and actively drives each such pixel. Simultaneously, as indicated at blocks D' through H', these basis steps are performed for light rays propagating along pathways r_Lj extending from point P_j in the spatial scene to the pupil of the driver's right eye. The specifics of such subprocesses will be described below.

As indicated at block D, the computer formulates in coordinate system R_c for each point P_j, the 3-D line equation r_ej extending from point P_j to point O_Le. The computer also formulates the 3-D surface equation S_LCLV for the principal x-y plane of coordinate system R_c. As indicated at block E, the computer system then computes coordinates x_Lqj, y_Lqj of the intersection point Q_Lj on the x-y plane of coordinate system R_c. This can be achieved by equating 3-D line equation r_ej and 3-D plane equation S_LCLV, and then evaluating the resulting expression where z=0. As indicated at block F, the computer then checks to determine whether the computed intersection point Q_Lj lies within the light intensity reducing boundaries of the LCLV surface. This is achieved using the following test criteria: 0≦|X_Lqj |≦w/2 and -H≦Y_Lqj ≦0.

Using computer coordinates X_eqj and Y_Lqj, the computer o then computes the address values A(x_Lqj, y_Lqj) at each intersection point Q_Lj. In the ideal case where each pixel has a very high resolution, there generally will be a one-to-one correspondence between a particular coordinate pair and a particular pixel. In many applications, however, pixel resolution will be less then ideal and a number of clustered coordinates can be assigned to a particular pixel on the LCLV surface. Such relationships between coordinate pairs and pixels can be represented in a table which is stored in memory and readily accessible by the computer.

Finally, as indicated in step H, controller/driver 16 uses address values A(x_Lqj, y_Lqj) and intensity reduction data m(P_j) to address and actively drive the particular pixels of the LCLV panel, thereby selectively reducing the intensity of incident light rays propagating along line r_Lj. As steps D through H are being performed, steps D' through H' are also performed for points P_j in the imaged scene. In this way computed pixels through which light rays r_rj intersect are electrically addressed and actively driven so as to reduce the intensity of light rays propagating towards the driver's right eye.

Referring to FIGS. 6 through 8B, the second illustrated embodiment of the electro-optical system of the present invention, will be described.

As illustrated in FIG. 6, the electro-optical element of system 20 is in the form of an optically reflective LCLV panel 21. In this illustrated embodiment, reflective LCLV panel 21 is adapted for pivotal mounting inside an automobile at a position presently occupied by conventional rear-view mirrors above the dashboard. It may, however, be installed outside the automobile vehicle and function as a side view mirror. Preferably, the width of the reflective LCLV panel is about 30 centimeters, the height thereof about 10 millimeters and the pixel size about 2.5 centimeters, although these dimensions will understandably vary from embodiment to embodiment. In construction, LCLV panel 21 differs from LCLV panel 8 in that a light reflective coating 22 is applied on the rear of optically transparent surface 22 so that the LCLV panel will function like a mirror.

On opposite sides of the upper portion of LCLV panel 23, a pair of camera stations 24A and 24B similar to camera SO stations 13A and 13B in FIG. 2, are mounted to provide a stereo scene-imaging camera subsystem which images spatial scenery within the driver's field of view through the rear window of the automobile. In order to determine the position of the pupils of the driver's left and right eyes, a pair of infra-red camera substations 25A and 25B, similar to camera substations 9A and 9B in FIG. 2, are mounted along a rotatable platform 26 to form a pupil-tracking camera subsystem. To perform the data processing functions described in connection with the system of FIG. 2, electro-optical system 20 also includes a computer system 27 and an LCLV controller/driver 28 operably associated as shown in FIG. 6.

In FIG. 7, a ray optics model is presented for electro-optical system of FIG. 6. As shown, coordinate system R₁, specified by x, y and z coordinate axes, is embedded within the reflective LCLV panel so that (i) the x-y plane of R₁ lies within the planar surface of the LCLV panel, (ii) the x axis is perpendicular to the optical axis of camera stations 24A and 24B, and (iii) the z axis is parallel with the optical axis of these camera stations. Coordinate system R₂, specified by x', y' and z' coordinate axes, is embedded within rotatable platform 26 so that (i) the x" axis is perpendicular to the optical axis of camera stations 25A and 25B, (ii) the y" axis is aligned with the y' axis of coordinate system R₁, and (iii) the center of origins of coordinate systems R₁ and R₂ spatially coincide. In order to measure the degree of rotation that coordinate system R₁ has undergone with respect to coordinate system R₂ at any particular positioning of the pupil-tracking camera subsystem, an optical encoder (not shown) is provided at the rotational axis between platform 26 and reflective LCLV panel 23. Data from the optical encoder regarding the relative rotation between coordinate systems R₁ and R₂ is provided to computer system 27 for processing in a manner to be described hereinafter.

Disposed at the center of the pupil of the driver's left eye is the origin of coordinate system R_LE which is specified by axes x_Le, y_Le and z_Le, with the z_Le axis aligned with the optical axis of the driver's left eye. Disposed at the center of the pupil of the driver's right eye is the origin of coordinate system R_RE which is specified by axes x_re, y_re and z_re with the z_re, axis aligned with the optical lD axis of the driver's right eye and the x_re axis aligned with the x" axis of coordinate system R_Le. As shown, the position of origin O_Le of coordinate system R_LE is specified by position vector t_L defined in coordinate system R_c. The origin O_Re of coordinate system R_RE is specified by position vector t_r also defined in coordinate system R_c. Points P_i in the spatial scene and virtual points P_i * behind the reflective LCLV panel are both specified by x_i, y_i, and z_i coordinates in coordinate system R_c. Virtual points P_i * in the imaged scene having an intensity above a predetermined threshold are designated as Pixel and their position in coordinate system R_c is specified by coordinates x_j, y_j, and z_j.

A light ray propagating from a point P_j in the spatial scene and passing through the pixel surface of LCLV panel 23 to point Q_Lj (enroute to the driver's left pupil) is specified by a 3-D line equation, r_1Lj. The path of the light ray reflecting off the reflective layer at point Q_Lj, towards the pupil of the driver's left eye, is specified by a 3-D line equation, r_2Lj. In accordance with the laws of physics, the angle of incident α₁ of the incident light ray is equal to the angle of reflection α₂ when measured from the normal vector n_Lj at point Q_Lj. Similarly, a light ray propagating from point P_j in the scene and passing through the pixel surface of LCLV panel to point Q_rj (en route to the driver's right pupil) is specified by a 3-D line equation, r_rj. The path of the light ray reflecting off the reflective layer at point Q_rj, towards the pupil of the driver's right eye, is specified by a 3-D line equation, r_2rj. At point Q_rj, the angle of incidence α₁ is equal angle of reflection oz when measured from the normal vector n_rj at point Q_rj.

As the driver views a spatial scene of upcoming traffic through the automobile rear window (i.e., by perceiving light reflected off the reflective LCLV panel), a number of steps are repeatedly performed by the intelligent electro-optical system. In this way, while the driver operates the automobile, light reducing pixel locations are automatically determined, electrically addressed and actively driven on a real-time basis reducing glare associated with upcoming headlights.

In FIGS. 8A and 8B, a generalized process is described for automatically reducing the intensity of light rays reflecting off the reflective layer of the LCLV panel, towards the driver's eyes. As indicated at blocks A, B and C, coordinate data regarding points P_i in the scene is gathered, the data processed to detect high intensity points P_j in the viewed scene, and the position of the driver's pupils are determined in a manner similarly described in connection with the system of FIG. 2. Then the sequence of steps D through H and D' through H' are performed in a parallel fashion in order to compute the x and y coordinates of the intersection points Q_Lj and Q_rj, which coincide with each high intensity point P_j in the viewed scene. Using this coordinate data, the corresponding pixels are electrically addressed and actively controlled to reduce the intensity of incident light rays, thereby eliminating glare.

Referring to FIGS. 9 through 11C, the third illustrated embodiment of the electro-optical system of the present invention will be described.

As illustrated in FIG. 9, the electro-optical element ID of system 30 is in the form of an optically transparent LCLV panel 31 adapted for pivotal mounting inside an automobile above the dashboard in a manner similar to a sunvisor. The LCLV panel has a width W, a length L and a rectangular border 32 surrounding the pixels of the panel. In night time glare reduction applications, pixel border 32 can be a luminous band realized by an optical fiber inlaid within a channel formed about the pixel perimeter. When light, from a source within the LCLV panel, is transmitted through the optical fiber, a rectangular band of low-level illumination is produced therefrom. In day time glare reduction applications, a black rectangular shaped pixel border, disposed against a reflective backing, can be provided for visually indicating the boundaries of the pixelled surface.

In order to acquire image(s) of oncoming traffic as seen by the driver along his or her field of view through the optically transparent LCLV panel, the driver wears a scene-imaging camera subsystem 33 embodied, for example, in a camera support frame 34 resembling the frame of a conventional pair of eyeglasses. As illustrated, the camera support frame has a bridge portion 34A with a nose bearing surface 34B and a pair of stems 34C and 34D which are adapted to encircle the driver's ears. In this embodiment, the scene-imaging camera subsystem comprises a pair of camera stations 33A and 33B disposed at opposite ends of the bridge portion. Acquired image data from the CCD image detection array in each camera station is transmitted through stems 34C and 34D to a computer 35 for processing in a manner described hereinabove. As will be described in greater detail hereinafter, the primary function of the computer system is to compute the addresses of pixels which are to be electrically addressed and actively driven so as to reduce the intensity of incident light rays from intense sources of illumination that have been detected in the acquired images of the scene being viewed by the automobile driver through the LCLV panel.

In FIG,. 10 a ray optics model is presented for the electro-optical system of FIG. 9. As shown, coordinate system R_c is embedded within the camera support frame so that (i) the x axis of is perpendicular to the optical axis of each camera station 33A and 33B, and (ii) the z axis is parallel with the optical axis of the camera stations. Disposed at the center of the pupil of the driver's left eye is the origin of coordinate system R_LE, which is specified by axes x', y' and z', with the z' axis aligned with the optical axis of the driver's left eye. Disposed at the center of the pupil of the driver's right eye is the origin of coordinate system R_RE which is specified by axis x", y" and z", with the z" axis aligned with the optical axis of the drivers right eye, and the x" axis aligned with the x' axis of coordinate system R_LE. The position of the origins of coordinate systems R_Le and R_re, relative to coordinate system R_c, will vary slightly from one automobile driver to another, but can be simply determined and related by a homogenous transformation T. Determining transformation matrix T can be achieved by a calibration procedure, much like fitting a pair of eyeglasses to a person's head. The procedure involves (i) determining the coordinates of points O_ie and O_re (relative to coordinate system R_c) using stereo image acquisition and analysis described above, and (ii) using such coordinates to compute transformation matrices for these coordinate systems.

As illustrated in FIG. 1, coordinate system R_LCLV, specified by principal axes x"', y" and z" , is embedded within the LCLV panel so that the x-y plane of R_LCLV lies within the pixel plane of the LCLV panel, and (ii) the center or origin of the coordinate system R_LCLV coincides with the point at which the left vertical pixel border intersects with the lower horizontal pixel border. Points P_i in the scene are specified by coordinates x_i, y_i, z_i measured in coordinate system R_c, and points in the scene having an intensity above a predetermined threshold are designated as P_j. The position of points P_j are specified by coordinates x_j, y_j, z_j measured in coordinate system R_c.

As the automobile driver views oncoming traffic through the automobile front window, a number of steps are repeatedly performed by intelligent electro-optical system 50, permitting the driver to operate his vehicle while light reducing pixel locations are automatically determined and electrically addressed and actively driven on a real-time basis.

In FIGS. 11A through 11C, a generalized process is described for automatically reducing the intensity of light rays which propagate through the optically transparent surface of the LCLV panel, towards the driver's eyes. As indicated at block A of FIG. 11A, the camera system acquires and buffers a pair of stereo images of the spatial scene viewed by the automobile driver through the LCLV panel. The acquired stereo image pair is then processed to determine the coordinates (x_i, y_i, z_i) for each point P_i represented in the image(s) of the spatial scene. As indicated at block B, the computer determines for each point P_i whether the detected light intensity is above a preselected intensity threshold, and if so assigns to each such point P_j a desired intensity reduction m(P_j).

As indicated as Block C, the computer then processes once again the acquired stereo image pair in order to recognize at least three selected points on the LCLV panel, namely P_A, P_B and P_C which lie on luminous pixel border 32. As illustrated in FIG. 10, P_A is selected as the point of intersection between left vertical and bottom horizontal pixel border lines; P_B is selected as the point of intersection between left vertical and top horizontal pixel border lines; and P_C is selected as the point of intersection between right vertical and bottom horizontal pixel border lines. As indicated at B block D, the computer uses these recognized points in the acquired stereo image pair to compute the coordinates of recognized points P_A, P_B and P_C, specifically: (x_A, y_A, z_A), (x_b, y_B, z_B) and x_C, y_C, z_C). Thereafter, as indicated in Block E, the computer formulates the equation for the principal x-y plane of coordinate system R_LCLV, with respect to coordinate system R_C. Then using the coordinates of the recognized points P_A, P_B and P_C, the computer computes the coefficients of the transformation matrix T for coordinate conversion from R_C to R_LCLV, and also the coefficients of the inverse transformation matrix T^-1 for coordinate conversion from R_LCLV to R_C.

At this stage of the process, the system performs in parallel the steps indicated at Block G through L for left eye ray optics, and Blocks G' through L' for right eye ray optics. Specifically, at Block G the computer formulates (in coordinate system R_c) for each point P_j, the equation of the 3-D lines r_Lj from point P_j to origin point O_Le, and the intersection point Q_Lj at which lines r_Lj intersects the plane or surface of the LCLV panel. At Block H, the computer computes the coordinates (x_Lj, y_Lj) of intersection point Q_Lj by equating the equations for r_Lj and the x-y plane of coordinate system R_LCLV. At Block I, the computer converts coordinates x_Lj, y_Lj from coordinate system R_c to R_LCLV using the inverse transformation matrix T-1. At Block J, the computer determines whether or not the coordinates of intersection point Q_Lj lie within the pixel boundaries of the LCLV panel, i.e., whether or not 0≦ X_Lqj ≦W/2 and 0≦Y_Lqj ≦h for all values of j in the acquired stereo image pair. Then, as indicated at Block K, the computer uses coordinates x_Lqj, y_Lqj to compute address value(s) A(x_Lqj, y_Lqj) at each ray-plane intersection point Q_Lj. Then to complete the process, the controller/driver addresses the pixels with pixel address value(s) A(x_Lqj, y_Lqj) and drives each addressed pixel with its desired intensity reduction m(P_j), thereby selectively reducing the intensity of incident light rays propagating along path r_Lj. When the steps indicated at Blocks G' through L' are carried out, intensity reduction of light rays propagating along path r_rj is achieved.

Referring to FIGS. 12 through 14C, the fourth illustrated embodiment of the electro-optical system of the present invention, will be described.

As illustrated in FIG. 12, the electro-optical element of system 40 is in the form of an optically reflective LCLV panel. This LCLV panel can be adapted for pivotal mounting, for example, inside an automobile at a position presently occupied by the rear-view mirror above the dashboard, or alternatively, outside the automobile as a side view mirror. The construction of reflective LCLV panel 41 is similar to the panel described in the second illustrative embodiment. In addition to an optically transparent LCLV surface 42 and reflective surface 43, reflective LCLV panel 41 has a rectangular shaped luminous border 44, as described in connection with the second illustrative embodiment above. All other components of the electro-optical system are similar to that depicted in connection with the third illustrative embodiment of the present invention.

In FIG. 13, a ray optics model is presented for the electro-optical system of FIG. 12. In nearly all respects, this embodiment of the electro-optical system is modelled identical to the electro-optical system of the third embodiment of the present invention. The only difference is that the image of points P_i and P_j (designated as P_i * and P_j * respectively) are used to formulate the path that light rays transverse from their respect position in the spatial scene, towards the pupils of the left and right eyes of the automobile driver. Notably, this ray optics approach takes advantage of a particular relationship which holds true for a planar reflective surface, namely: that the distance from intersection point Q_Lj to point P_j * equals the distance from point Q_Lj to point P_j, and likewise the distance from point P_i to point Q_rj equals the distance from point P_i * to point Q_rj. Consequently, virtual points P_i * in acquired images of a spatial scene can be used in the process illustrated in FIGS. 11A through 11C. This process is depicted in FIGS. 14A through 14C, and is identical to the process of FIGS. 11A through 11C in all other respects.

In each of the four illustrative embodiments of the present invention described above, a general approach has been provided to the problem of modulating the amplitude of incident light rays using electrically addressable pixelled surface having controllable light transmittance. While these approaches provide a general solution to the problem over any range of system operations or configurations, there will be applications in which system constraints permit the use of computing techniques less complex in the computational sense.

FIG. 15, a ray optics model is presented for a fifth embodiment of the electro-optical system of the invention. As illustrated, the system is generally indicated by reference numeral 50. In this embodiment of the present invention, the automobile driver wears a single camera station 51 supported by a support frame similar to the one illustrated in FIG. 10. The camera station is operably connected to a computer which is interfaced with a controller/driver similar to that shown in FIG. 10. A coordinate system R_c, specified by principal axes x, y and z, is embedded in the camera station so that (i) the z axis is aligned with the optical axis of the camera station, and (ii) the origin of coordinate R_c system is positioned at the principal plane of the image forming optics of the camera station.

Disposed at the center of the pupil of the driver's left eye is the origin of coordinate system R_LE which is specified by axes x', y' and z' with the z' axis aligned with the optical axis of the driver's left eye. Disposed at the center of the pupil of the driver's right eye is the origin of coordinate system R_RE which is specified by axis x", y", and z", with the z', axis aligned with the optical axis of the driver's right eye, and the x" axis aligned with the x' axis of coordinate system R_LE.

The position and orientation of the origins of coordinate systems R_c and R_LCLV are determined using a position and orientation tracking system well known in the art. Using such position and orientation data, a homogeneous transformation between coordinate systems R_c and R_LCLV can be computed and used to convert the coordinates specified in coordinate system R_c to coordinates specified in coordinate system R_LCLV.

When using the electro-optical system of FIG. 15 in automotive applications, the distance S measured between the LCLV panel and the driver's pupils will be about 0.3 meters, and points of intense illumination P_j will typically reside at a distances of 2 or more meters from the driver's pupils. In such cases, the ratio of x/h will be 0.9 or greater, indicating that incident light rays propagating through the LCLV panel to the driver's pupils will be approximately parallel. As will be described below, this condition permits simplification of the computing approach used in determining the ray-panel intersection coordinates, L_j and R_j.

The simplified process involves first acquiring an image of the spatial scene within the driver's field of view using the camera system. The pixels of the acquired image are processed to detect intensity levels above a preselected threshold. Then, the x, y coordinates of each image pixel corresponding with point P_j (having an intensity above the threshold) are converted from coordinate system R_c to the coordinate system R_LCLV, embedded in the LCLV panel. This coordinate conversion step is achieved using the homogeneous transformation matrix computed above. The condition x/h≧0.9 justifies this step since the x, y coordinates of each intersection point Q_j can be equated with the pixel coordinates on the camera's image detection plane, while introducing only a minor degree of error. From the coordinates of point Q_j, the coordinates of intersection points L_j and R_j can be computed using coordinate geometry and the average measure of interpupil distance, h≈0.065 meters. Having computed the coordinates of L_j and R_j, the computer then computes the addresses A(_Lj) and A(_Ri) of corresponding pixels allowing for a sufficiently low pixel resolution in order to compensate for errors in L_j and R_j. Using computed addresses A(L_j ) and A(R_j) and intensity reduction data m(P_j), the controller/driver addresses particular pixels and actively drives them to change their light transmittance to the desired value.

The above-described approximation technique for computing ray-plane intersection points L_j and R_j can be applied to reflection-type electro-optical systems as well.

Having described the illustrative embodiments of the present invention, several modifications are contemplated.

For example, the LCLV surface of the system hereof can be fashioned to the geometry of a windshield and/or rear window of an automobile, while achieving the objects of the present invention.

A portable electro-optical system is envisioned, in which a monocular or stereo camera subsystem is embodied within a head supported frame having a pair of LCLV lenses, each disposed in front of one of the wearer's eyes. The resulting device can be worn as a pair of eyeglasses to block out solar glare during the day, and headlight glare at night.

The electro-optical system of the present invention may be installed within unmanned vehicles so as to protect image detecting components used in navigational systems. The electro-optical system of the present invention may operate at wavelengths within or outside the optical spectrum.

The present invention has been illustrated in applications in which glare produced by man-made illumination sources has been eliminated. However, glare and noise produced by solar sources can also be effectively reduced using the principles of the present invention.

While the particular embodiments shown and described above will be useful in many applications in the glare reduction art, further modifications of the present invention herein disclosed will occur to persons skilled in the art to which the present invention pertains. All such modifications are deemed to be within the scope and spirit of the present invention defined by the appended claims.