PROJECTOR WITH PHASE HOLOGRAM MODULATOR
20260099123 ยท 2026-04-09
Inventors
- Muralidhar Madabhushi Balaji (Dallas, TX, US)
- Jeffrey Matthew Kempf (Dallas, TX)
- Terry Alan Bartlett (Dallas, TX, US)
Cpc classification
G03H1/0443
PHYSICS
H04N9/3102
ELECTRICITY
G03H1/0866
PHYSICS
G03H2210/10
PHYSICS
G03H1/2294
PHYSICS
G03H2225/24
PHYSICS
G03H1/08
PHYSICS
G03H1/14
PHYSICS
International classification
G03H1/22
PHYSICS
G03H1/08
PHYSICS
G03H1/14
PHYSICS
Abstract
An example projection system includes a processing element to run an iterative algorithm to generate, based on a digital video image, a phase hologram; an illumination source to output coherent light; a phase light modulator (PLM) to receive the phase hologram and the coherent light, and modulate the coherent light to produce modulated light; and projection optics to receive the modulated light from the PLM and to project an image responsive to the modulated light and the phase hologram, in which the image has more noise in bright regions of the image than in dark regions of the image, and in which the bright regions are of higher intensity than the dark regions.
Claims
1. A projection system comprising: a processing element configurable to run an iterative algorithm to generate, based on a digital video image, a phase hologram; an illumination source configurable to output coherent light; a phase light modulator (PLM) configurable to receive the phase hologram and the coherent light, and modulate the coherent light to produce modulated light; and projection optics configurable to receive the modulated light from the PLM and to project an image responsive to the modulated light and the phase hologram, in which the image has more noise in bright regions of the image than in dark regions of the image, the bright regions being of higher intensity than the dark regions.
2. The projection system of claim 1, wherein the iterative algorithm is executable to generate the phase hologram such that, when used to project the image by the projection optics, the phase hologram allows noise into the bright regions of the image and inhibits noise from entering the dark regions of the image.
3. The projection system of claim 1, wherein the PLM includes an array of individually addressable storage cells and respective micromirrors configurable to direct the modulated light to the projection optics.
4. The projection system of claim 1, wherein the PLM includes a liquid crystal on silicon (LCOS) PLM.
5. The projection system of claim 4, wherein the LCOS PLM includes liquid crystals sealed between glass and semiconductor layers with aluminum pixel electrodes.
6. The projection system of claim 1, wherein the processing element is further configurable to run the iterative algorithm to: produce an initial phase hologram in a source plane associated with the PLM; and iteratively perform forward and inverse Fourier transforms between the source plane and a target plane where the image is projected, in which weighted amplitude updates that preserve zero amplitude in the dark regions while allowing noise to accumulate in the bright regions are applied.
7. The projection system of claim 6, wherein the processing element is further configurable to terminate the iterations when an error between a generated image in the target plane and a goal image falls below a threshold.
8. The projection system of claim 1, wherein the processing element is further configurable to run the iterative algorithm to: produce an initial phase hologram in a source plane associated with the PLM; perform a Fourier transform on the initial phase hologram to generate, in a target plane where the image is projected, a target plane image having an amplitude term and a phase term; update the amplitude term of the target plane image to generate an updated transformed expression for an updated image; and perform an inverse Fourier transform on the updated transformed expression to generate in the source plane a first iteration phase hologram indicated by a first iteration phase term.
9. The projection system of claim 1, wherein the processing element is further configurable to run the iterative algorithm k times, k being a positive integer that begins at 1 and is incremented for each iteration, as follows: produce a k.sup.th phase hologram in a source plane associated with the PLM; perform a Fourier transform on the k.sup.th phase hologram to generate, in a target plane where the image is projected, a k.sup.th target plane image having a k.sup.th amplitude term and a k.sup.th phase term; update the k.sup.th amplitude term of the k.sup.th target plane image to generate a k.sup.th updated transformed expression for a k.sup.th updated image; and perform an inverse Fourier transform on the k.sup.th updated transformed expression to generate in the source plane a (k+1).sup.th phase hologram in the source plane for a next iteration.
10. The projection system of claim 9, wherein: the k.sup.th phase hologram is expressed as e.sup.j.sup.
11. The projection system of claim 10, wherein the processing element is further configurable to terminate the k iterations when an error between a last generated target plane image and the goal image falls below a threshold.
12. The projection system of claim 1, wherein: the projection optics is configurable to receive the modulated light from the PLM and to project multiple different images responsive to the modulated light; and the processing element is configurable to: generate multiple different phase holograms for the multiple different images using different initial phase holograms, respectively; and control the PLM to sequentially display the multiple different phase holograms with the multiple different images, respectively.
13. A method comprising: produce an initial phase hologram in a source plane associated with a phase light modulator (PLM); iteratively performing forward and inverse Fourier transforms between the source plane and a target plane where an image is generated, including applying weighted amplitude updates that preserve zero amplitude in dark regions of the image while allowing noise to accumulate in bright regions of the image, and generating an updated phase hologram after each iteration; and terminating the iterations when an error between the generated image in the target plane and a goal image falls below a threshold, in which a most recent updated phase hologram is a final phase hologram.
14. The method of claim 13, wherein the final phase hologram appears as a randomized pattern at phase light modulator (PLM) and produces a recognizable image when projected.
15. The method of claim 13, wherein the image is derived from an input digital video signal.
16. The method of claim 15, wherein the method produces multiple initial phase holograms for the image derived from the input digital video signal, generates, after each iteration, multiple updated phase holograms for the multiple initial phase holograms, respectively, and produces multiple final phase holograms.
17. The method of claim 16, further comprising: sequentially displaying the multiple final phase holograms with the image.
18. The method of claim 17, wherein the multiple final phase holograms are sequentially displayed with the image using time averaging.
19. A projection system comprising: multiple color illumination sources including red, green, and blue lasers; a phase light modulator (PLM) to sequentially display color-specific phase holograms; and a processing element to generate separate phase holograms for each color channel to preferentially distribute noise to bright regions of a projected image and inhibit noise from entering dark regions of the projected image.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0003] For a more complete understanding of the illustrative examples of aspects of the present application that are described herein and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
[0004]
[0005]
[0006]
[0007]
[0008]
[0009]
[0010]
[0011]
[0012]
[0013]
[0014] Corresponding numerals and symbols in the different figures generally refer to corresponding parts unless otherwise indicated. The figures are drawn to clearly illustrate the relevant aspects of the illustrative example arrangements and are not necessarily drawn to scale.
DETAILED DESCRIPTION
[0015] The making and using of example arrangements that incorporate aspects of the present application are discussed in detail below. It should be appreciated, however, that the examples disclosed provide many applicable inventive concepts that may be embodied in a wide variety of specific contexts. The specific examples and arrangements discussed are illustrative of specific ways to make and use the various arrangements, and the examples described do not limit either the scope of the specification, or the scope of the appended claims.
[0016] For example, when the term coupled is used herein to describe the relationships between elements, the term as used in the specification and the appended claims is to be interpreted broadly, and is not to be limited to connected or directly connected but instead the term coupled may include connections made with intervening elements, and additional elements and various connections may be used between any elements that are coupled. The term optically coupled is used herein. Elements that are optically coupled have an optical connection between the elements but various intervening elements may be between elements that are optically coupled.
[0017] The term phase hologram is used herein. A phase hologram is a pattern containing phase information that is generated for display on a phase light modulator when amplitude is made constant. The phase hologram is transmitted to a phase light modulator as a two dimensional bitmap. An array of picture elements (pixels) of the phase light modulator move in correspondence with the bitmap, which is stored in storage cells associated with the pixels. The pixels phase modulate illuminating light.
[0018] The terms bright regions and dark regions are used herein to describe portions of a projected image. As used herein, a bright region of a projected image is a region with greater intensity, where more light is visible, and the bright regions are regions that have an intensity at least 50% of the maximum intensity, while the maximum brightness that could be found in a bright region is where the maximum possible amplitude of the light is projected. A dark region in a projected image is a region with lesser intensity, a dark region is one with intensity of less than 50% of the maximum intensity. The maximum darkness that could be found in a dark region is where the minimum possible amplitude of the light is projected, which could be zero. In some examples, the image intensity is binary, so that the goal amplitude for a particular pixel in the projected image is either 0 (dark region) or 1 (bright region). The term noise is used herein. As used herein, noise is a variance in intensity of light in a region in a projected image. The noise is introduced, in part, by algorithms used to generate a phase hologram that is used to modulate illumination light, and the modulated light is used to project an image. In the arrangements, noise that occurs in projecting an image from a phase light modulator is distributed in the bright regions (amplitude is 1 in the binary case) of a projected image, and noise is reduced or eliminated from the dark regions (amplitude is zero in the binary case) of the projected image. The noise reduction in the dark regions is achieved by the methods used to generate phase holograms used in the arrangements. The images projected using the arrangements have high contrast, as the noise is excluded or minimized in the dark regions, increasing the contrast.
[0019] Digital projectors using spatial light modulators (SLMs) are increasingly used. Examples of SLMs include projectors using liquid crystal on silicon (LCOS) light modulators and digital micromirror device (DMD) light modulators. Applications for spatial light modulators include heads up displays (HUD), cinema, television, presentation projectors, headlights, near eye displays, light detection and ranging (LIDAR), automotive console displays, forward looking infrared (FLIR), structured light, spectroscopy, 3D printing, light field displays, communications, and high dynamic range projectors. In some digital projectors and displays, amplitude modulation is used with SLMs. In amplitude modulation, SLMs display patterns corresponding to an image, and an illumination source provides light to illuminate the pattern on the SLM. Modulated light is reflected from the SLM to projection optics which projects an image on a screen or a surface for viewing. Amplitude modulation using SLMs is subtractive. To obtain variable light intensity in the projected images, the SLM either reflects all, or less than all, of the illumination light to the projected image. The light that is not reflected for projection is reflected instead at another direction away from the projection optics and into a light dump. For color images, multiple color light sources are used to illuminate different corresponding images displayed on the SLM in a sequence in the different colors that are projected during a frame display time. Because the human viewing system integrates the light, the viewed image is a color image containing all of the colors.
[0020] Images may be projected at a screen, onto the human retina, or for viewing in a far field using phase light modulators (PLMs). Using phase light modulation, a modulated wavefront is produced in the light that is projected towards a screen or to another far field distance for viewing. Phase information displayed on the phase light modulator is a hologram. Due to constructive and destructive interference of light in the traveling wavefront as it traverses the path from the phase light modulator, the projected image appears at the screen, on the human retina, or at a selected far field distance for viewing. PLMs are optically efficient when compared to amplitude modulators, because all of the illumination light is directed for projection, and no light dump is used for any of the illumination light. The PLM device displays phase holograms, which may appear as a randomized pattern when viewed at the PLM, but when the PLM is illuminated, phase modulated light is projected to the screen or at a far field distance, and the projected image appears for viewing due to constructive and destructive interference.
[0021] PLMs have physical characteristics that result in noise in a projected image. The contrast ratio of the projected image is limited in part because the phase modulator has a physical bit depth (number of bits per pixel corresponding to physical operation of the pixels of the phase modulator). Using the phase light modulator to modulate light introduces quantization noise, due to a fixed number of positions of the pixels on the phase modulator. Additional noise results when an optimization approach is used to generate phase holograms. The phase holograms are generated in an algorithm that starts from a random initial phase image, which may introduce some noise. The amplitude in the phase plane is fixed, which also introduces noise. These characteristics of projectors using PLMs result in noise in the projected images, lowering the contrast ratio. It is advantageous to increase the contrast ratio in the projected images.
[0022] To use a phase light modulator in a projector, phase holograms are generated for display on the phase light modulator. The phase holograms, when illuminated, cause a projected image to appear at a target surface or target plane. Intensity in the image varies due to constructive and destructive interference of the phase modulated light waves reflected from the PLM. The projection of the modulated light from the PLM to the screen or far field imaging plane corresponds to a Fourier transform between a source plane (where the PLM is located) and a target plane (where the projected image appears). The projection path from a phase representation (intensity or amplitude is uniform) at the source plane (where the PLM device is located) transforms the phase hologram to a transformed amplitude and phase image representation corresponding to the projected image in the target plane.
[0023] A projector system using a PLM generates phase holograms from amplitude images. The amplitude images may be provided as digital video images retrieved from a memory, received from a computer executing an internet browser, received from a video streaming device, video player, camera, camcorder, game console, or other image source. An algorithm is used for generating the phase holograms to provide bitmap patterns to the PLM. An iterative algorithm that uses Fourier transforms to determine phase information for an image is described in a paper titled A Practical Algorithm for the Determination of Phase from Image and Diffraction Pictures, R. W. Gerchberg and W. O. Saxton, Cavendish Laboratory, Cambridge, United Kingdom (1971), which is hereby incorporated by reference herein in its entirety. This algorithm is referred to as the G-S algorithm. The G-S algorithm uses an iterative loop to optimize the phase information so that the transformed image obtained by the algorithm is within an error threshold from the image to be reproduced. In the G-S algorithm, the phase information is updated at each iteration, and the projected image is optimized repeatedly in an iterative loop, until the transformed image that results from the generated phase information is within an error threshold from the image to be reproduced.
[0024] The phase information algorithm has inherent noise due to the fact that in the source plane of the transforms, where the phase representation information is present, the amplitude is fixed. The reconstructed image in the target plane becomes a convolution of the image with the Fourier transform of the inverse of the amplitude in the phase plane, which is a source of noise in the projected images. Further, some noise is introduced by the algorithm's use of a random initial phase image. The random initial phase image can be obtained using a pseudo-random pattern generator, for example. During the optimization process, the G-S algorithm distributes noise in both the dark and bright regions of the projected image as the algorithm iteratively moves the transformed image towards the image to be reproduced by repeatedly modifying the phase information. The presence of this noise, particularly the visible noise in the dark regions of the projected image, reduces the image contrast. The noise can be expressed as the mean of the intensity in a region, and as variance in the intensity of the region. When noise increases, both the intensity mean and the intensity variance increase. When noise decreases, both the intensity mean and the intensity variance in a region decreases.
[0025] An approach to improving the contrast in images projected using phase modulation is to allow the amplitude to vary, increasing the number of free variables in the reconstruction plane. In some approaches, regions of the projected image are identified as don't care areas where both amplitude and phase may take any arbitrary value. By reducing the number of pixels that are used for the projected image, the number of pixels that need to be accurately represented is reduced, and the noise in the viewable image regions may be reduced. The cost for using these approaches is a reduction in the field of view. In an example referred to as the Fidoc algorithm, the algorithm puts a don't care band around the image, reducing the size of the image, and the Fidoc algorithm directs the noise into the unused band. While effective in increasing the contrast in the viewed images, the Fidoc algorithm, and other similar approaches, results in less usable field of view, as the usable field of view is reduced by the unused band area.
[0026] In the arrangements, an iterative algorithm generates phase holograms that result in increased contrast in projected images. In the algorithm, noise in the projected image is allowed in bright regions of the projected image and noise is reduced or eliminated in the dark regions of the projected image. The intensity mean and the intensity variance in the dark regions is reduced (compared to the G-S approach without the arrangements); while the intensity variance and intensity mean in the bright regions is allowed to increase, and both values are greater in the brighter regions than in the dark regions. The algorithm of the arrangements generates phase holograms which, when used to project an image, use the entire field of view in the projected image. To further reduce noise in the bright regions, in additional arrangements time averaging may be used, where a number of images are sequentially projected using different phase holograms generated by the algorithm. By projecting these images sequentially with varying noise distributions in the bright regions, the noise perceived in the bright regions of the projected images is further reduced. In this way, high contrast images can be produced using the arrangements. Because contrast is a ratio of the intensity in the bright regions to the intensity in the dark regions, as the noise is reduced or eliminated from the dark regions, the dark region intensity decreases, as does the mean of the intensity and the variance of the intensity in the dark regions, increasing the contrast.
[0027] In the arrangements, an illumination source provides coherent illumination light. The coherent illumination light is optically coupled to a PLM. The PLM displays phase holograms. The PLM phase modulates the illumination light. The phase modulated light is directed from the PLM to projection optics. In the projection optics, a Fourier optical element optically couples the modulated light from the PLM to an output for the projector. The projected image corresponds to a Fourier transform from the phase hologram displayed on the PLM. For a projected image, an iterative algorithm generates a phase hologram for display on the PLM. In an iterative process for generating the phase hologram, the algorithm weighs the dark regions of the projected target plane image heavier than the bright regions, allowing noise (increased intensity variance and increased mean intensity) in the bright regions of the projected image while reducing or excluding it in the dark regions. The phase holograms determined by the algorithm result in increased contrast in the projected images. The noise that is present in the bright regions of the projected image may be temporally averaged, and therefore reduced, by projecting the image using several different phase holograms in rapid succession. This approach averages random noise in the bright regions, so that the perceived noise in the projected image is reduced. In one example arrangement, a MEMS PLM is used. The MEMS device has a high rate of operation, making temporal averaging in the bright regions of the projected images expedient. In an alternative arrangement, an LCOS PLM is used.
[0028]
[0029] In an example, the PLM 105 is a MEMS PLM device. The MEMS PLM includes an array of addressable storage elements associated with individual MEMS micromirrors that have multiple vertical positions (vertical with respect to a reflective surface of the device when the device is facing upwards). The micromirrors may be used to modulate the phase of illumination light and, when illuminated, output phase modulated light. In the illustrated example, the PLM 105 is reflective and the phase modulated light is directed to projection optics 106. The projection optics 106 is arranged to display an image 107 at a viewing distance, for example by projecting the image onto a screen, onto a human retina, onto a display medium, a wall or at a far field in a field of view. The relationship between the phase hologram at the PLM and the projected image may be expressed as Fourier transforms, and the projection optics 106 is sometimes referred to as a Fourier lens. The projected image 107 has amplitude and phase even though the bitmap displayed on the PLM is a phase hologram. The phase hologram may appear as a random bitmap or as noise when viewed on the PLM. The intensity observed in the projected image resembles a natural image to the human vision system, despite the stochastic nature of the phase hologram displayed on the PLM. Alternative illumination sources may be used, such as a super luminescent diode. For a color projection system, multiple color light sources may be used to illuminate phase holograms displayed by the PLM in a sequence to provide a color image. In an example red, green and blue lasers may be used to independently illuminate the PLM in a sequence of sub frames while the PLM displays phase holograms for each color. The projected image is then a color image.
[0030] As shown in
[0031]
[0032]
[0033] In operation, a MEMS PLM device, which has thousands, hundreds of thousands, or even more micromirror pixel elements arranged in rows and columns, stores patterns in individually addressable storage cells arranged in an array to correspond with the pixels. Bitmap patterns for phase holograms may be loaded into the storage cells from a device controller. The PLM is then updated to display a phase hologram based on the stored pattern. The PLM is used to phase modulate illumination light to produce the projected image to be viewed in the target plane. This operation is repeated for each projected image. Pulse width modulation (PWM) of the illumination source may be used with the PLM to display sub-frame images in a sequence to further improve the projected image, reduce heat in the system, and reduce stress on the pixels. Multiple colors can be used with sub-frame images of differing weight (longer or shorter display times) to produce color projected images by using multiple color illumination sources, for example red, green and blue color sources, to form color projected images.
[0034]
[0035]
[0036] In operation, phase holograms are generated for images to be projected for display on the PLM. Using these phase holograms to modulate the illumination light creates projected images output by the projector system (see 500 in
[0037] As described above, one prior approach to generating phase information for images is to use the G-S algorithm. The G-S algorithm performs an iterative approach that takes advantage of the fact that a change in either the amplitude or phase in one domain of the Fourier transform, for example a source plane corresponding to where the PLM is located, results in changing both amplitude and phase distributions in the opposite domain of the Fourier transform, for example the target plane where a projected image is shown. Therefore changes in the phase hologram in a source plane (corresponding to where the PLM modulates illumination light) may affect changes in both amplitude and phase in the image at a target plane (where a screen or other display plane receives the projected image). In this way, a phase hologram for a projected image may be determined, and phase light modulators may produce arbitrary images projected in the target plane. More than one phase hologram may be used to project a given target image. In the iterative G-S algorithm a random phase hologram is used as a starting point in the phase plane. Differences in the random starting phase information may result in different solutions for the phase hologram for a given target image. The G-S algorithm error (a difference between an image generated using the phase information from the algorithm, and the target image) decreases with iterations, and so the algorithm converges.
[0038] In the G-S algorithm, a random initial phase hologram is generated for the first iteration of the algorithm. This can be done using a pseudo-random pattern generator, for example. For the index k, where k is a positive integer starting at 1 and incrementing by 1 for each iteration, the phase term e.sup.j.sup.
[0039] If, after an iteration, the error threshold that is determined by comparing the amplitude image output by the algorithm after k iterations to the goal image is below a threshold, the algorithm ends. The final phase term e.sup.j.sup.
[0040] The G-S algorithm is an optimization algorithm. In the G-S algorithm, the amplitude of the image obtained from the algorithm is made to be within an error threshold of the goal image intensity. Noise in the projected image produced by the G-S algorithm is distributed in both the dark and bright regions of the projected image. This noise, which is caused in part by the variable phase with uniform amplitude nature of the source plane information in the G-S optimization, results in reduced contrast in the projected image. In some applications, for example heads up displays (HUDs), high contrast is very advantageous. Improvement in the contrast ratio obtained in the projected image is beneficial, particularly for these applications.
[0041]
[0042]
[0043] At step 717, the amplitude term is updated. In the arrangements, the expression used to update the amplitude is:
[0045] At step 713, the inverse Fourier transform is performed, and this returns the source plane expression with amplitude and phase {square root over (I.sub.source,k)}e.sup.j.sup.
[0046] The algorithm shown in
[0047]
[0048] The phase hologram generation in step 801 may be done before the rest of the steps of
[0049]
[0050] At step 903, the iterative algorithm starts by performing a Fourier transform to obtain the transformed amplitude and phase terms {square root over (I.sub.target,k)}e.sup.j.sup.
[0051] At step 907, as described above, the amplitude term of the transformed image is updated as described by the equation:
At step 909, the updated amplitude is used with the phase term, and an inverse Fourier transform is performed to return the expression to the source plane, so the expression with amplitude and phase is:
At step 911, the index k is incremented, and the algorithm returns to step 903, the Fourier transform using the updated phase hologram to begin the next iteration.
[0052] The algorithm continues iteratively, updating the phase hologram in the source plane, performing Fourier transform on the source plane phase hologram to a transformed phase and amplitude term, and checking the error in the amplitude by comparing to the goal image amplitude and comparing the error to an error threshold, until the error is less than the error threshold. When the algorithm ends at step 912, the final phase hologram is ready for use in projecting the goal image.
[0053]
[0054] The phase light modulator PLM 105 may be an LCOS SLM. PLM 105 may also be implemented using a MEMS PLM device. The PLMs have individually addressable pixels that modulate phase of illumination light, and output the phase modulated light.
[0055] Although the example illustrative arrangements have been described in detail, it should be understood that various changes, substitutions and alterations may be made herein without departing from the spirit and scope of the present application as defined by the appended claims. For example, where a single laser diode was shown, multiple laser diodes or an array of laser diodes may be used. Accordingly, the appended claims are intended to include within their scope processes, machines, manufacture, compositions of matter, means, methods, or steps that provide equivalents to the examples disclosed.