Television – Video display – Projection device
Reexamination Certificate
1999-07-22
2002-08-13
Brier, Jeffery (Department: 2672)
Television
Video display
Projection device
C348S806000
Reexamination Certificate
active
06433840
ABSTRACT:
TECHNICAL FIELD
This invention relates generally to calibrating the deflection in a vector or raster image projector, and more particularly to correction of distortion in projection display systems which use cathode ray tubes (CRTs) to project raster images.
BACKGROUND ART
In a projection display system using cathode ray tubes (CRTs), a computer image generator produces simulated high resolution raster images which are projected onto a projection surface. When a raster image is projected using a CRT projection system, the image is often distorted. The sources of image distortion include but are not limited to: off axis projection and viewing; non-planar display surfaces (e.g. domes or curved screens); and nonlinear errors in the electronics, optics and magnetic devices in the system. The correction of the image for distortion sources is commonly performed by distorting the CRT's raster shape.
Historical methods of adjusting the CRT raster have included linear corrections, non-linear corrections, or digital simulation of-linear and non-linear corrections. Other methods of error correction are based-on interpolated surface correction values including, values organized as a matrix of bilinear interpolation values or spline surface error correction.
For each of these methods, the user adjusts a set of control parameters to make the image appear correct for a given view location, projector placement, display surface and the current state of the projection system.
The historical methods of adjusting analog electronic circuits to correct for image distortions have now been replaced by digital methods. The analog circuits required careful adjustments and had problems of stability and drift common to analog electronics.
The use of digital simulations of the analog corrections provided a system which had the stability of digital methods and separated the correction electronics from the user controls. Since the analog electronics attempted to correct each of the sources of error with a separate circuit, the early digital simulations performed this same modeling.
It was found that the modeling of each possible error source was cumbersome. First, there is the problem of identifying all of the potential sources of error. Then there is the problem of creating a suitable set of error correction equations. This results in needing to adjust many error functions. Geometric errors introduced by the projection onto a screen and viewed from a different location than the projection source, require a different model for each projection surface shape. It is also difficult to know which combination of error functions needs to be adjusted to calibrate the CRT's raster for an observed visual distortion. This required an iterative calibration technique since there was an interaction between the collection of functions. It also allowed a technician to “work into a corner” which would require them to reset the controls and start over.
Interpolated surface error correction is a much more general solution. Interpolated surfaces do not attempt to model each error source. Rather, it is a method of reacting to total errors as a single adjustable function in two dimensions. The surface control points can be generated in a number of ways. Originally the digital simulations of the analog correction models were used to set the control point values. Direct user adjustments of the control point values was later introduced.
The control surface to interpolate for the correction values can be implemented using any number of interpolation methods. For ease of hardware implementation of the interpolation, one method currently used in the art is bilinear interpolation. For bilinear interpolation to appear smooth, a very large number of control points must be used. It is cumbersome and time consuming for the user to adjust all of the bilinear control points directly. The current state of the art allows a user to directly adjust the control points for a bicubic spline surface which is interpolated to generate the bilinear control point values used by the hardware. With recent hardware advances it would now be possible to directly implement the bicubic spline surface in hardware and drop the step of using the bilinear surface. Bicubic spline surfaces have the advantage of being second order continuous and thus do not introduce the discontinuities introduced by piecewise bilinear surfaces. It is thus possible to use far fewer control points.
When using the spline surface correction, the user adjusts a set of control points
18
, as shown in
FIG. 1
, to make them match up with a set of alignment points. The alignment points represent the correct location of the control points
18
for a properly calibrated system. The user must adjust all of the control points
18
on the spline surface. For the surface to be easy to calibrate, the number of points to adjust must be kept to a minimum. In direct opposition to the desire to have a minimum number of control points, larger numbers of control points provide more localized adjustment which is a capability that may be necessary to provide the required calibration accuracy. It is best not to add too many control points for two reasons. First, the time to adjust the system increases with the number of control points. Second, extra control points increase the order of the correction surface with the potential of creating wiggles in the correction.
FIG. 2
illustrates a misplaced point anomaly
19
that can be introduced by individual control point adjustments (i.e a wiggle). Another limitation, in using the spline surfaces is apparent when looking at the physics of the system. From a mathematical view of the spline surface, any amount of adjustment can be performed with each control point being totally independent in placement. In reality, control point adjustment is not independent of the physical drive limitations in the electronics and magnetic components of the image projector. A large adjustment in one point will cause other control points to appear to move or drift.
Accordingly, it would be an improvement over the state of the art to provide a new method and apparatus for adjusting control points in a control grid where the adjustments do not introduce misplaced points, wiggles or drift through the correction process.
It would also be an improvement in the art to provide an effective method to adjust control points in a control grid which more effectively represents the physical drive limitations in the electronics and magnetic components of the image projector.
OBJTECTS AND SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method and apparatus for adjusting control points for control grids in an image display where the adjustments to the control points do not introduce misplaced points.
It is another object of the invention to provide a method and apparatus for adjusting control points in control grids which saves calibration time.
It is another object of the invention to provide a method and apparatus for adjusting control points in control grids which minimizes the problem of raster drift.
It is another object of the invention to provide a method and apparatus for adjusting controls points in control grids using an intermediate delta surface which is then added to the resulting surface shape.
It is yet another object of the invention to provide a method and apparatus for adjusting control points in control grids using bicubic interpolation between the adjusted points and the original surface shape.
The present invention is realized in a method and apparatus for multi-level image adjustment. The method for multi-level image alignment has a base grid with control points. The first step in the method is creating a delta surface with a plurality of control points equal to the number of control points in the base grid. In addition, the delta surface control point values are set to zero. The next step is defining a control grid with a desired number of control points from the base grid. Then an adjustment value is entered for a selected control
Brier Jeffery
Evans & Sutherland Computer Corporation
Thorpe North & Western
Yang Ryan
LandOfFree
Method and apparatus for multi-level image alignment does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for multi-level image alignment, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for multi-level image alignment will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2950555