Television – Video display – Projection device
Reexamination Certificate
2000-11-02
2004-05-18
Lee, Michael H. (Department: 2614)
Television
Video display
Projection device
C348S750000, C348S756000, C359S279000, C359S237000, C398S188000
Reexamination Certificate
active
06738105
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field of the Invention
The present invention relates to removal of low-frequency pattern content from coherent light to avoid the perception of speckling.
2. Background Art
A coherent light beam will exhibit interference fringes if there are plural propagation paths that the light can take from the light source to the display device. Imaging systems use focusing elements such as lenses, mirrors, and the like to direct light onto a display device. Each of these lenses etc. may offer a vast number of propagation paths for the light to reach any given display surface location or pixel. This problem is very analogous to the familiar dual-slit wave/particle interference experiments from basic physics. In a two-dimensional display, these interference paths are exhibited as speckles, which resemble bright spots or “snow”, noise which is familiar from poor broadcast television reception.
The human eye has a flicker fusion rate of approximately 50 Hz to 60 Hz. Speckles which appear in different locations for sufficiently short periods of time will not be perceived as moving by the human eye, as the nervous system will fuse the flickering values at those particular retinal locations into a single distribution of perceived values. As the time-averaged speckle image will have low average surface brightness the result will be a dim background image that is easily ignored.
The human eye is very sensitive to coherent structures that vary in time. A simple example is a tiger moving through tall grass. If the tiger remains stationary, our eye has a difficult time discerning the tiger, due to his camouflaged pattern. But if the tiger is moving, our eye will discern the systematic disturbance of the pattern, even if the grass is, itself, moving in a strong wind—it is not the movement which is perceived in this case, it is the structural disturbance in which a low-frequency pattern (the tiger's stripes) is injected into the overall image.
A problem, similar to reducing coherent light speckling, occurs in the field of halftone printing. Halftone printing represents a continuous gray-scale image via a fixed-grid pattern of purely black, fixed sized dots. Ideally, the eye will perceive the intended gray level by fusing the dots and white background into a lower resolution retinal image. Many half-toning algorithms are known, but they suffer from varying degrees of artifacting, such as banding, in which the eye perceives extra structure in the dots. The eye is very sensitive to such structural features. Patterns can be removed from our perception by removing coherent, low-spatial frequency structure from the source.
Early half-toning algorithms used white noise to randomize the dots. White noise contains generally equal content from low frequencies, medium frequencies, and high frequencies. White noise can be filtered to emphasize or eliminate certain components. Blue noise contains more high-frequency content than medium-frequency and low-frequency content; by way of contrast, pink noise emphasizes the lower frequencies. For purposes of this patent, blue noise should be understood to mean noise which has sufficiently little low-frequency content that it is not interesting to or not perceived by the human eye or, more generically, by the video perception apparatus of the intended audience, whether human or machine. It may be thought of as “anti-pattern”.
In the field of half-tone printing, researchers have discussed using blue noise to overcome banding in halftone printing; Robert A. Ulichney, “Dithering with blue noise”, Proc. EEE, vol. 76, no. 1, p. 56, January 1988; and Theophano Mitsa and Kevin J Parker, “Digital halftoning using a blue noise mask”, ICASSP, 91: 1991 International Conference on Acoustics, Speech, and Signal Processing, pages 2809-2812, Toronto, Canada, May 1991 (WEEE).
FIG. 1
shows a despeckling system used in a video display. Light from a coherent light source was passed through a phase shift plate, and the phase shift plate was mechanically rotated. To achieve despeckling, the phase shift plate needed to have varying phase shift characteristics at different portions of its surface, and the plate needed to be rotated fast enough that any given such portion of its surface was within one display pixel's worth of the light beam for a period of time less than the flicker fusion time of the human eye.
There are some problems with this solution. For one, the perimeter of the plate has a greater rotational velocity than the center of the plate; thus, portions of the image that pass through the plate near its center may receive less despeckling than portions which pass through the perimeter. That portion of the light which passes through the rotational center of the plate will receive essentially zero despeckling; in order to avoid that, it would be necessary to use a plate having a diameter more than twice the diameter of the light beam at that location—so the entire image could be passed through the plate off to the side of the rotational center—which results in a larger overall display system, a larger and more expensive plate, and larger and more expensive rotation mechanism. Another alternative might be to oscillate the rotating plate so it does not have a consistent rotational center, but that would introduce undesirable mechanical problems such as vibration.
Neither the halftone printing art nor this prior art despeckling system makes any mention or use of blue noise characteristics for despeckling. Indeed, the distribution and characteristics of the phase shifting characteristics of respective regions of the rotating plate are not described in terms of even white noise (with its undesirable low frequency pink noise component), much less blue noise.
REFERENCES:
patent: 3652956 (1972-03-01), Pinnow et al.
patent: 3891300 (1975-06-01), Tsunoda
patent: 3917380 (1975-11-01), Kato et al.
patent: 4783753 (1988-11-01), Crimmins
patent: 5093563 (1992-03-01), Small et al.
patent: 5272473 (1993-12-01), Thompson et al.
patent: 5398130 (1995-03-01), Redman
patent: 5416628 (1995-05-01), Betti et al.
patent: 5459532 (1995-10-01), Baba
patent: 5566382 (1996-10-01), Worchesky et al.
patent: 5748311 (1998-05-01), Hamann et al.
patent: 5892607 (1999-04-01), Atlas
patent: 5912755 (1999-06-01), Bergano
patent: 6154259 (2000-11-01), Hargis et al.
patent: 6191880 (2001-02-01), Schuster
patent: 6252693 (2001-06-01), Blauvelt
patent: 6323984 (2001-11-01), Trisnadi
patent: 6424412 (2002-07-01), Matthews
patent: 6490071 (2002-12-01), Logan, Jr. et al.
Toshiaki Iwai et al., Speckle Reduction in Coherent Information Processing, May 1996, Processing of the IEEE, vol. 84, No. 5.*
Theophano Mitsa, Kevin J. Parker; “Digital Halftoning Using A Blue Noise Mask”, IEEE, 1991, pp. 2809-2812, USA.
Robert A. Ulichney, “Dithering with Blue Noise”, Proceedings of the IEEE Jan. 1988, pp. 56-79, vol. 76, No. 1, USA.
Hannah Eric C.
O'Connor Michael
Desir Jean W.
Lee Michael H.
Trop Pruner & Hu P.C.
LandOfFree
Coherent light despeckling does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Coherent light despeckling, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Coherent light despeckling will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3189756