WO1995018423A1

WO1995018423A1 - Spatial stabilization for bandwise imaging device

Info

Publication number: WO1995018423A1
Application number: PCT/US1994/014085
Authority: WO
Inventors: David J. Schoon
Original assignee: Schoonscan, Inc.
Priority date: 1993-12-29
Filing date: 1994-12-07
Publication date: 1995-07-06
Also published as: AU1431995A

Abstract

An electro-optic imager, used for exposing photosensitive media, wherein a plurality of imaging beams is used for creating an image. Imaging is done in bands. Imaging and cross-scan movement alternate in time. During the cross-scan movement times, an optical interrupter (202) is preferably used to sense a ruling (114) to determine coarse positional error. Then a second fine-tuning optical element may be used to compensate for residual error. Once this is completed, imaging resumes with the next band. The coarse errors may be characterized and program data modified so as to minimize these errors on subsequent accesses to the same band positions.

Description

SPATIAL STABILIZATION FOR BANDWISE IMAGING DEVICE
Field of the Invention
The present invention relates to imaging devices, and more particularly to imaging devices where an image is recorded on photosensitive media by a plurality of light beams.

Background of the Invention
Imagers, usually laser imagers, are used to record electronically produced information on photographically sensitive media (e.g. photographic film or paper; or photoconductive drums, as in laser printers). Such devices are used in applications wherein information exists in an electronic form and a visually observable presentation of the information is desired. Typically a single beam of light is directed sequentially across a photosensitive media, in a raster scan fashion.

Typically a laser (e.g. a gas laser beam) is used to create the beam.

Such imagers are used to create images when a computer is used to control or create the image information. Examples of the functions which are best done with computer assistance are: (1) size adjustment and cropping; (2) combinations of same or other photos into a layout; (3) addition of text or other graphics to photos; (4) color corrections (e.g. correction for use of daylight film with incandescent lighting); (5) unsharp masking (i.e. electronic sharpening of a photograph); (6) darkness and contrast adjustments; and (7) retouching. Hardcopy images are also needed when printing plates are being-pre-pared, and a proof of the electronic image information must be obtained prior to preparation of printing plates from that same image information. Another example of needs for electronic imagers is in the preparation of medical x-rays (e.g.

from CAT [Computerized Axial Tomography] or NMR [Nuclear
Magnetic Resonance] scans).

Laser imagers usually image with a single beam of light. In order to obtain acceptable overall speeds a high relative speed of the beam with respect to the media is necessary. One prior design involves wrapping the photosensitive media around the outside of a drum.

The drum is spun rapidly while the optical position of the beam is advanced slowly and continuously with a leadscrew.

Another common design involves using a rapidly spinning polygon mirror which deflects the light onto the photosensitive media. The media is usually moved slowly past the optical system in a direction perpendicular to the direction of scan.

Both prior systems require that the rapidly moving component (the drum in the first case, or the polygon mirror in the second case) move with little if any variation in speed. If this is not the case image elements which should line up from scan line to scan line might not do so, but rather appear nonuniform.

Industry jargon calls these nonuniformities "jaggies" since lines or object boundaries which should appear smooth appear instead jagged. A second consequence of nonuniform motion would be-undesired variations in opacity, especially if the images which are being created are continuous tone images.

It is also required that the slowly moving component (the light beam in the first case, or the media in the second case) move uniformly. If this is not the case, undesired variations in opacity will result. More specifically, increased exposure will result when the scan lines are too close together and decreased exposure will result when the scan lines are too far apart. These conditions would result from media motion which is too slow or too fast, respectively.

Because the motion is slower and thus has a smaller inertial effect relative to frictional effects, the more slowly moving element is usually more difficult to control adequately.

In order to minimize these imperfections, various electronic controls and high mechanical precision are used in various components. Considerable cost is involved both in these controls and in the high mechanical precision. Consequently, a need exists in the art for a low cost yet high precision system for controlling components in an imaging system to reduce variations and discontinuities and produce high quality images therewith.

Summary of the Invention
The present invention addresses these and other problems associated with the prior art in providing optical scanning of a photosensitive media wherein the spacing between the scan lines is extremely uniform (i.e., not having significant variations in spacing between various scan lines), so that discontinuities are minimized in the scanned images. Further, the invention provides these high quality images with a comparatively low cost of manufacturing, not requiring expensive machining of mechanical components or expensive electronic controls as has heretofore generally been required.

The invention is preferably used in an imaging system in which a plurality of light beams is used for imaging (e.g. 20-200 beams). In such a system, the photosensitive media is wrapped around the outside of a rotating drum (or cylinder), but the drum is rotated more slowly than is generally feasible with prior art devices. The imaging occurs in discrete bands with no movement of the optical system during the imaging process.

In accordance with one aspect of the invention, there is provided an imaging apparatus for forming an image on photosensitive media. In the apparatus, an imaging means applies a plurality of light beams to a photosensitive media. A first media transport means moves the photosensitive media and the imaging means relative to one another in a first direction, and a second media transport means moves the photosensitive media and the imaging means relative to one another between a plurality of imaging bands defined along a second direction that is generally orthogonal to the first direction. Further, a control means operates the imaging means in a first period to apply the plurality of light beams to the photosensitive media when the photosensitive media and the imaging means are at one of the plurality of imaging bands such that an image band is exposed on the photosensitive media.

In a second period, the control means actuates the second media transport means to move the photosensitive media and the imaging means relative to one another in the second direction to a relational position proximate one of the plurality of imaging bands. A positioning means is utilized to position the photosensitive media and the imaging means relative one another in the second direction at a discrete relational position corresponding to one of the plurality of imaging bands such that uniform spacing is provided between adjacent image bands exposed on the photosensitive media.

In accordance with another aspect of the invention, there is provided a spatial stabilization apparatus for use in a bandwise imaging system of the type having imaging device, a portion of which is configured to apply a band of light beams to a photosensitive media as the photosensitive media moves relative to the portion of the imaging system in a scanning direction, and a media transport mechanism for moving the photosensitive media and the portion of the imaging device relative to one another between a plurality of imaging bands in a cross-scanning direction which is generally orthogonal to the scanning direction. In the apparatus, a band positioning controller is operable during a dead cycle in which scanning is disabled on the bandwise imaging system.

The band positioning controller actuates the media transport mechanism to move the photosensitive media and the portion of the imaging device relative to one another in the cross-scanning direction to a relational position proximate one of the imaging bands.

The band positioning controller also positions the photosensitive media and the portion of the imaging device relative to one another in the cross-scanning direction at a discrete relational position corresponding to one of the imaging bands such that uniform spacing is provided between adjacent imaging bands exposed on the photosensitive media.

In accordance with a further aspect of the invention, there is provided a method for spatially stabilizing in a bandwise imaging system of the type having scanning means for applying a band of light beams to a photosensitive media as the photosensitive media moves relative to the scanning means in a scanning direction, and media transport means for moving the photosensitive media and the scanning means relative to one another between a plurality of imaging bands in a cross-scanning direction which is generally orthogonal to the scanning direction. In the method, the media transport means is actuated during a dead cycle in which scanning is disabled on the bandwise imaging system to move the photosensitive media and the scanning means relative to one another in the cross-scanning direction to a relational position proximate one of the imaging bands.

Also, the photosensitive media and the scanning means are positioned relative to one another in the cross-scanning direction at a discrete relational position corresponding to one of the imaging bands such that uniform spacing is provided between adjacent image bands exposed on the photosensitive media.

In a preferred embodiment, a portion of the optical system is moved between discrete positions at times when no imaging occurs. The exact position of this optical system is sensed by an optical interrupter which senses an optical ruling. Once the optical system has come to rest after being moved, the error in its position is determined by the optical interrupter. This error information is used to adjust a secondary optical component (a fine-tuning element). This secondary optical component corrects for errors in the position of the first optical component. This occurs before the light beams are restored and imaging resumes. The system alternates between imaging (with the optical system being stationary) and repositioning for the next band of image (with the light beams turned off).

In an alternative embodiment, the exact positioning of a portion of the optical system is assured by use of a pin and groove system which secures the portion of the optical system at discrete, reproducible points along the cross-scanning direction. Further, in another embodiment errors in the positioning may be detected and used to provide correction in imaging data and/or in the imaging position to reduce imperfections in the image.

These and other advantages and features, which characterize the invention are pointed out with particularity in the claims annexed hereto and forming a part hereof. However, for a better understanding of the invention, its advantages and objectives attained by its use, reference should be made to the drawing which forms a further part hereof and to the accompanying descriptive matter, in which there is described a preferred embodiment of the invention.

Brief Description of the Drawing
FIGURE 1 is a plan view of a preferred imaging system consistent with the present invention.

FIGURE 2 is a perspective view of the stationary linear ruling and traveling optical interrupter of
Figure 1, the interrupter moving with the objective lens assembly.

FIGURE 3 is a graph which shows the relationship between the voltage output from the optical interrupter in Figure 2 and the interrupter position with respect to the linear ruling.

FIGURE 4 is a block diagram of a control system consistent with the present invention for operating the imaging system of Figure 1, as well as the primary objects connected to that control system.

FIGURE 5 is a plan view of an alternate imaging system consistent with the present invention.

FIGURE 6 is a cross-sectional view of the objective lens assembly of Figure 5, taken along line 6-6.

FIGURE 7 is a cross-sectional view of the mirror assembly of Figure 5, taken along line 7-7.

FIGURE 8 is a partial enlarged side elevational view of the pin and groove engagement system of Figure 5.

FIGURE 9 is an intensity profile resulting from a central band of image correctly placed between two other bands.

FIGURE 10 is an intensity profile resulting from a central band of image slightly misaligned between two other bands.

FIGURE 11 is an intensity profile resulting from intensity compensation for band misalignment consistent with the principles of the invention.

FIGURE 12 is a block diagram of a portion of the electronic circuitry which adjusts overlapping intensities for the imaging system of Figure 5.

Detailed Description of the Preferred Embodiment
My issued U. S. patents 5,054,893 and 5,225,851, and co-pending applications 07/827,061 and 07/884,408 (filed on January 28, 1992 and May 19, 1992, respectively), discuss various methods of imaging with multiple parallel beams of light. To the extent they are required to support the disclosure, the disclosures of all of these references are incorporated by reference herein. In summary, according to these patents and applications, a plurality (e.g., 64) of light beams are generated using lanthanum-modified lead zirconate titanate (PLZT) electro-optic light modulators to control more or less collimated light. By writing to a photosensitive media with multiple beams, a high overall speed (square inches of image generated per second) is obtained even with a small relative speed between any beam and the media.

Because this speed is small the images can be created in bands, with the imaging system alternating between (1) imaging a band (e.g., 0.2" wide), with no movement of the optical system in the cross-scan direction, and (2) movement of the optical system in the cross-scan direction, with no imaging occurring.

Band Positioning
Turning to the figures, wherein like numbers denote like parts throughout the several views, Figure r shows a preferred imaging system design 10 consistent with the present invention. A drum or right circular cylinder, 101, is supported on shafts 102 and 103.

Photographically sensitive media (e.g. Kodak Ektacolor
RA photographic paper or Ektachrome film) is wrapped around the drum, emulsion side out, and held to the drum with clamps appropriate for the media. A first media transport means 20 moves the photosensitive media and an imaging means 100 relative to one another in a first, or scanning, direction, which is the direction in which successive light beams are scanned bandwise onto the media. In the preferred embodiment, first media transport means 20 is a motor which imparts a rotation to drum 101; however, in other embodiments, the media may be oriented in a plane, in which the relative movement would then be lateral as opposed to rotational.

Other orientations known in the art may be used in the alternative.

Imaging means 100 is now described which controls the deposition of light beams onto the photosensitive media in accordance with a data flow of image data.

Objective lens assembly 104, which contains first surface mirror 1041 and lens 1042, slides on guide rods (not shown) and is moved by leadscrew 106. This leadscrew is connected to a timing pulley which, via a timing belt, is turned via a stepping motor (not shown), the general operation of which is known in the art.

This mechanical configuration operates to provide a second media transport means which moves the photosensitive media and the imaging means relative to one another in a second, or cross-scanning, direction which is generally orthogonal to the scanning direction.

Due to the bandwise scanning of the device, a number of imaging bands are thus defined along this direction.

Mirror assembly 105, which contains first surface mirrors 1051 and 1052, also slides on guide rods (not shown) and is moved by the second media transport means through leadscrew 107. This leadscrew is also connected to a timing pulley and to the same timing belt and stepping motor as used to rotate leadscrew 106. The timing pulleys are chosen so that leadscrew 106 rotates two turns for every one turn of leadscrew 107. The two leadscrews 106 and 107 are of the same pitch. Because of this arrangement mirror assembly 105 moves half as much as objective lens assembly 104 for any given rotation of the stepping motor connected to both assemblies.

The following items are stationary and do not move: lamp 108, lens 109, filter 116, light valve 110, and aperture array 111. Glass plate 112 rotates about shaft 113. The shaft does not move translationally, but does rotate about its longitudinal axis.

Lamp 108 provides light which is collected and more or less collimated by lens 109. The brightest spot of lamp 108 is focused onto objective lens 1042. The light which passes through lens 109 is filtered by filter 116, which preferably is a wheel of which one of several filter colors is selected. For color film and paper, one each of red, green, and blue filters is provided on that wheel.

Light valve array 110 is an array of light valves (e.g., two rows of 32 cells per row) together with polarizers before and after the light valve chips. The polarizers are oriented so that with no voltage on any cell the polarization vector is not rotated within the
PLZT chip and little if any light passes through the second polarizer (which is crossed with respect to the first polarizer). However, if a voltage is applied to an electrode for any light valve cell, light passes in accordance with the amount of polarization rotation, up to a maximum of light when a 900 rotation occurs.

Typically this maximum is about 27% of the light incident on the first polarizer.

Aperture array 111 controls the beam energy profile of each beam. My copending application 07/884,408 discusses the way these beam energy profiles are set so as to minimize the appearance of scan lines on the output. This array is preferably one of many such arrays, chosen for the particular media and resolution selected. For some resolutions an interleaving of scan lines is used.

Glass plate 112 is rotated on shaft 113 so as to microposition the image slightly. Both of my copending applications 07/827,061 and 07/884,408 discuss the way in which a rotated glass plate can microposition the image. Using Snell's law it can be easily calculated that a 1.5mm thick glass plate with index of refraction = 1.523 will move the image approximately 0.00068" for every degree of rotation.

The array of apertures is focused by lens 1042 onto drum 101. Dotted line 115 shows the optical path. The total optical distance from apertures 111 to lens 1042 is always constant since assembly 105 moves half as much as assembly 104.

In one embodiment the following specific parts are used, though those skilled in the art will recognize that many other possibilities could be used. The lamp is an Optical Radiation Corporation XM150-lHS 150-watt short arc xenon lamp which produces over half of its illumination in a sphere less than 1 mm in diameter.

Lens 109 is actually an array of short focal length lenses which image this spot onto objective lens 1042.

Filter wheel 116 contains three dichroic filters for red, green, or blue light, and is rotated by a stepping motor. Light valve array 110 is two rows of light valves, each 32 cells on 0.030" centers, the two rows 0.6" apart. The light valves are PLZT material, such as is described in my issued U. S. Patent No. 5,054,893.

The glass plate is rotated by a stepping motor and gear train (30:1), with coiled spring on the output shaft to eliminate gear backlash. The optical distance from aperture array 111 to objective lens 1042 is 16.24".

Lens 1042 has a focal length of 75 mm, and is optically 3.61" from the drum. The reduction ratio from apertures 111 to drum 101 is 4.5:1. Thus the bands are slightly over 0.2" wide. The 8.5" diameter drum turns at approximately two revolutions per second. The time required to create a 20" x 24" color print at 300 dpi is less than four minutes.

Overlapping of data is preferably used so that a gradual feathering of information can occur, between bands, more or less depending on the actual amount of mismatch between ideal and actual band separation. For 300 dpi images, the drum turns one revolution each for red, green, and blue filtering and data. For 600 dpi, the glass plate interleaves the data and two passes are made at each color. For 1200 dpi, four passes are made for each color, with four different positions of interleaving. For Ektacolor media, since the sensitivity to red light is much lower than it is to green and blue light, the number of rotations for red filtering and data may be more than one revolution, thus providing greater exposure in that wavelength.

It should be noted that the spacing between adjacent scan lines within a certain band is controlled by the aperture spacing and the magnification ratio, which in turn is determined by the positions of the optical components. The spacing between the last scan line in one band and the first scan line in the next band is controlled by the separation between bands. If this first separation from line to line is not virtually exactly the same as this second separation from line to line, a visible and objectionable discontinuity will result. Thus it is necessary that virtually no error in band to band positioning occur, that is, it is necessary that uniform spacing is provided between adjacent bands.

By "uniform", I mean that the band positioning be precise and reproducible between each of the adjacent bands to ensure proper positioning of all of the bands in an image. In embodiments of the present invention this positioning is primarily accomplished by actuating the second media transport means to move the photosensitive media and the imaging means relative to one another in the cross-scanning direction to another imaging band when scanning is disabled (i.e., in a dead cycle), then positioning the imaging means at a discrete relational position corresponding to the desired imaging band. In the preferred embodiment, this relational position is defined by the position of objective lens assembly 104 with respect to drum 101.

The exact position of objective lens assembly 104 is preferably sensed by a positional sensing means which includes an optical interrupter 202 attached to the bottom of assembly 104 and stationary ruling 114. These items are shown in greater detail in Figure 2. Lines are placed on the ruling along its longitudinal axis with a 0.2" period (0.1" line and 0.1" space).

Otherwise the ruling is optically transparent. Thus, alternating opaque and transparent portions are defined.

The interrupter 202 is an assembly with a light emitting diode on one side and a phototransistor on the other side. The sizes of each item are kept small so as to obtain a highly accurate sensing of the position of a light todark or dark to light transition. In one embodiment, the interrupter which is used is TRW
OPB813S7, which has a 0.007" aperture size.

In most positions the optical interrupter is totally off (viewing dark line area) or totally on (viewing light area of the ruling, 114). In the transitional regions, however, a changing voltage output from the interrupter (and appropriate driving circuitry) is obtained with very slight changes in position as illustrated in Figure 3. Ordinate 301 shows the voltage output of the positional signal provided by the interrupter, and abscissa 302 shows position of the moveable interrupter 202 with respect to the stationary ruling 114. Curve 303 shows the relationship between the two. For the aforementioned interrupter 202 the movement required to go from fully on to fully off is 0.007". Using this curve (with the "on" value being learned from the maximum that is obtained between lines) the positional error over a +/-0.0035" region can be determined.

With a 12-bit analog to digital converter sensing the output from this optical interrupter, the .007" total distance is sensed with an accuracy of approximately .007" divided by 4096, or 0.043 microns, which is more than adequate.

In a preferred embodiment, after a band of image has been created in a first period the cells are set in the off state, i.e., scanning is disabled.

Alternatively, a vane is placed between lamp 108 and lens 109 which rotates so as to block all light from reaching the aperture array. During a second period, preferably one rotation time of the drum, a stepping motor first rotates leadscrews 106 and 107 so as to obtain as precisely as possible a movement to the next band of image (0.2" in the example shown above) . The stepping motor used to move these leadscrews is microstepped, with one coil receiving more or less a sine wave, and the other coil receiving more or less a cosine wave. More specifically,
Phase A voltage = A sin(x)
Phase B voltage = A cos(x)
During movement, x is increased from its old value smoothly to a new value, x(i), where i is the band number. For a total image width of 20" and band widths of 0.2", there would be 100 possible band positions.

The exact discrete point of stopping (a "band position" representative of the discrete relational position of the imaging band) for each band is preferably "learned" via a characterization process during factory setup. During this characterization it is desired that the output from encoder 202 will be halfway between the light value and the dark value when the leadscrew comes to rest for each band. To the extent that this is not true, different data for the desired stopping point are stored in a memory or other storage device so that on subsequent attempts to stop at exactly a certain point a better set of data for x(i) is obtained. The exact reason for positional errors may not be known.

It might be imperfections in the leadscrew, nonlinearities in the stepping motor (sine and cosine functions might not be optimal, depending on the magnetic characteristics of the motor) imperfections in the timing pulleys or belt, or other causes. It is found that with this process, even with relatively inexpensive components, considerable accuracy can be obtained. With 50 characterization cycles, each cycle involving advancing to each of the 100 possible band locations, the positional error drops from typically +/-0.002" average error to typically +/-0.00007" (+/-1.8 microns).

The x(i) values which are learned are preferably modified slightly during subsequent user operation by a real-time characterization routine. The modified values are stored within a nonvolatile memory. Such updating of x(i) information compensates for wearing of parts, including wear of the leadscrew. The routine compares the positional signal from the optical interrupter with the positional value for the band position stored in memory (the "expected" value) and modifies the memory value accordingly.

The ruling 114 is produced by computer imagers (e.g. Gerber plotters) intended for high resolution film imaging. However, some imperfections will exist on the film. Once the imager of the present invention is completed, output images are scanned to determine errors in this ruling, thus characterizing the ruling.

Discontinuities in opacity of images at band boundaries are measured, and a relative set of better band placements determined. This is converted into interrupter voltage outputs which are desired at each band, these being other than the 50% voltage level used in the initial characterization. (Alternatively, the rotatable glass plate 113 position corresponding to a nominal leadscrew position may be varied.) By this process, the ruling 114 is also characterized.

The degree of precision obtained via leadscrew and ruling characterization is adequate for many applications. However, for other applications, typically high-resolution applications, greater accuracy may be needed. In a further embodiment, once the positional error of assembly 104 after a movement to a new band is known, a second step involves fine-tuning by the rotation of the glass plate 112. For the example given, each degree of rotation causes 0.00068" of image movement; the image reduction is 4.5:1; the stepping motor is half-stepped and has 200 steps/revolution, and is geared 30:1. Thus each step in the stepping motor results in 1/6000 revolution or 0.06 degree. This single step causes the image on the drum to move 0.23 microns, or 13% of the average inaccuracy of the leadscrew.

Further, as an additional correction, the intensities of the beams in the overlap region between bands may be adjusted. This procedure is discussed in greater detail below.

Therefore, one preferred embodiment of the present invention causes the following to happen in the time between imaging of bands:
1. The control system actuates a stepping motor to rotate the leadscrew and move the objective lens assembly to proximate a new band position.

2. The positioning system microsteps the stepping motor and stops the leadscrew at the position at which it has been "learned" the best band position is likely to occur. Typically, the positioning system retrieves the value of the band position from memory and actuates the transport mechanism accordingly until the positional signal from the positional sensor is substantially equal to a reference positional signal representative of the band position. The positioning is preferably performed as the leadscrew is nearing the band position during the movement of the objective lens assembly between bands, taking into account factors such as the deceleration of the assembly, and essentially determining the discrete stopping point for the assembly at the next band position.

Alternatively, the positioning may be performed after the assembly has been stopped at an approximate band location, whereby the assembly may then be microstepped in either direction to correct for positioning errors.

3. Once the leadscrew has come to rest, the positional error for that band is determined by measuring the encoder 202 output, as compared to the ideal expected output for that particular band. This positional error is used to calculate the amount of glass plate rotation which is required to best compensate for that error.

4. The glass plate is rotated to the best position to compensate for the residual error.

5. The amount of still residual error, even after glass plate movement, is calculated. This small residual error (which is generally less than 1/2 of 0.23 microns) is used to control the contour of beam energies in the overlap region between bands (discussed in greater detail below).

It should be understood that the glass plate position is additive to glass plate movements required to obtain interleaving of beams for 600 dpi and higher resolutions. It should also.be understood that the overlap contour used to adjust for residual positioning error is additive to overlap contours adjusted to compensate for out-of-round errors in the drum manufacture. These latter errors are compensated for in real time, based on the angular position of the drum.

Once the respositioning of the objective lens assembly 104 and glass plate 112 is completed, imaging is resumed. Normally this repositioning occurs within the time of one rotation of the drum (typically 0.5 second). Thus for typical imaging there maybe for each band one red cycle, one green cycle, one blue cycle, and one dead (or nonimaging) cycle, repositioning of the optical system occurring during the dead cycle.

The control system which accomplishes these objectives is illustrated in block diagram fashion in
Figure 4. Only the primary components of the imager are illustrated in this drawing. A microprocessor control system, 401, contains an interface (not shown) to the computer system supplying data. It also contains volatile and nonvolatile memory parts. This control system performs band selection and related fine-tuning, as well as imaging, by supplying control to the leadscrew stepping motor, 402, the lamp, 403, the glass plate stepping motor, 406, and high voltage drivers, 404, which control the light valves, 405. Control system 401 also controls a drum drive motor, 407, filter wheel stepping motor, 408, and aperture wheel stepping motor, 409. It receives data from optical interrupter 202.

The control system also preferably contains a means for insuring that the clocking of data is in concert with the rotation of the drum, this portion not being shown but being described in my copending application entitled "Clocking Means for Bandwise
Imaging Device", filed on an even date herewith.

Systems consistent with my invention should be contrasted with systems according to the prior art wherein imaging and optical system movement occur simultaneously. In such prior systems, if positional errors are detected it is too late to make any correction for same, since imaging has already occurred.

The speed at which errors develop is generally faster than mechanical corrections can be applied. Moreover, for continuously moving systems feedback means for detecting positional errors are more costly and less accurate. Still further, in prior art systems there is no opportunity to compensate for positional errors by adjusting an overlap region between scan lines. In contrast, systems consistent with the present invention are capable of detecting and compensating for positional errors during dead cycles so that those errors will be minimized as soon as imaging is resumed.

Alternative Embodiments
Those skilled in the art will recognize that many embodiments, other than the above preferred embodiments, are possible within the scope of this invention. For example, a piezoelectric movement of the apertures or one of the lenses may be used instead of the rotating glass plate. Alternatively, a very fine secondary leadscrew system may obtain this cross-scan micropositioning movement.

The leadscrew system for movement of the objective lens and mirror assemblies may be replaced by other systems, such as a wire rope or timing belt system.

Such systems might have higher inherent errors, but the fine adjust capability disclosed in this invention could render such systems effectively error-free, despite these errors.

Instead of using a rotating drum, another embodiment may involve a rotating polygon mirror, producing a moving bundle of beams of light. The optical system which produces these beams of light may be moved in the cross-scan direction, or the media may be moved relative to the beams of light. A ruling and optical interrupter would measure the exact positional error of each band. A rotating glass plate, a secondary leadscrew, or a piezoelectric element would accomplish fine-tuning of the band to band spacing. Imaging periods of time would alternate with periods of time involving translation in the cross-scan direction of one portion of the system relative to another portion.

Figures 5-8 show an alternative embodiment of the invention which utilizes a pin and groove engagement system for controlling bandwise positioning in the imaging system. As seen in Figure 5, for example, lamp 108 provides light which is approximately collimated by lens 109, similar to the embodiment shown in Figure 1.

Preferably, lens 109 forms an image of the lamp filament on lens 1042', for maximum optical efficiency. The light passing through lens 109 is directed through light valve array 110 which includes polarizer 110a, PLZT chip 110b and polarizer 110c in the manner discussed above.

Also, the light is filtered by color filter 116 and apertured by aperture array 111, which determines the size and shape of each beam of light. The light emerging from this aperture array is directed along optical path 115 by a number of components in the optical system, namely mirrors 1052' and 1051' on mirror assembly 105', and mirror 1041' and objective lens 1042' on objective lens assembly 104'. The light which passes through the lenses and mirrors in the optical system forms an image of apertures 111 onto the surface of drum 101.

Both assemblies 104' and 105' are at times permitted to translate (move) in a cross-scanning direction parallel to the drum's axis, on shaft 130.

Bushings 1045 and 1055 (best seen in Figures 6 and 7) within each of assemblies 104' and 105', respectively, permit this sliding motion. Such sliding motion of assemblies 104' and 105' is inhibited when pins 1043 and 1053, respectively, are permitted to fully engage grooves in a threaded rod 134.

Whether a pin engages threaded rod 134 or not is determined by the rotational position of a flattened shaft 132. When the flattened shaft is rotated causing the flattened portion to move to a position other than the one shown, a slider 1044 (shown in Figure 6) attached to assembly 104' causes the assembly to rise away from the threaded rod 134, so the pin 1043 will clear threaded rod 134, thus permitting translation of this assembly along smooth shaft 130 to occur. When the flattened portion of flattened shaft 132 is placed closest to slider 1044, pin 1043 engages threaded rod 134, and translation cannot occur.

The force which pushes pin 1043 into the threaded rod 134 is preferably caused by a wire rope system (not shown for clarity), which accomplishes the translating motion and provides the force required to hold the pin 1043 against one of the grooves in threaded rod 134.

A similar construction is used in assembly 105'.

As seen in Figure 7, pin 1053 is made to either contact, or not contact, a groove in threaded rod 134, being controlled by the rotation of flattened shaft 132 which contacts slider 1054. The same wire rope system as is used for assembly 104' is also used in the same manner to translate assembly 105' and provide the force required to hold the pin 1053 against one of the grooves in threaded rod 134.

The wire rope system which causes assemblies 104' and 105' to move, when permitted to do so, is attached to a capstan and a stepping motor (not shown for clarity). However, the mechanical advantages of the two assemblies are made to be different so that for every unit of movement of assembly 104', there is only half as much movement (in the same direction) of assembly 105'.

It can be seen that in this way the total optical path from the apertures 111 to objective lens 1042' is always held constant. Thus the image of apertures 111, on the drum 101, is always at the same (or very nearly the same) magnification ratio, and always in focus.

Threaded rod includes a plurality of precisely spaced grooves along its longitudinal axis. Since the preferred rod is threaded, the various grooves along its length are machined as a single continuous groove.

However, since pins 1043 and 1053 engage only along a narrow surface of threaded rod 134, one skilled in the art will appreciate that the grooves could be formed on a wide variety of shafts or surfaces. For example, grooves may be formed as a plurality of parallel rings along a rod, or may be formed only along the narrow surface of a rod to which the pins engage. Also, rather than a cylindrical rod, other types of shafts, plates, surfaces, etc. may be used to form the plurality of grooves.

The spacing between grooves in threaded rod 134 is made to be a 2x (or 2Nx, where N is an integer) integral multiple of the desired interval between bands of images. For example, if there are 32 PLZT cells, of which 30 are used, and the desired resolution on the drum is 300 dpi, the interval between bands is 30/300 = 0.100". The interval between grooves is 0.050", or 20 grooves per inch. Therefore, for a translation between two adjacent imaging bands, assembly 104' will be translated two grooves, and assembly 105' will be translated one groove.

The wire rope system positions the objective lens assembly 104' very close to the desired position for each band. After each band has been imaged, the flattened shaft 132 is turned by stepping motor 136 to permit translation. The stepping motor controlling the wire rope system moves both the objective lens and mirror assemblies by 0.1" and 0.050" respectively. This movement is approximate, inasmuch as the wire rope is a slightly elastic media. When this is completed, stepping motor 136 returns the flattened shaft to its original position wherein the pins 1043 and 1053 are dropped into new positions along threaded shaft 134.

The pins force each assembly 104' and 105' to find a position which corresponds to the exact shape and position of the associated groove. These positions are nearly totally reproducible, from time to time. As a result, the placement of each band of image on the drum is predictable, and the optical magnification ratio for each band is predictable.

As seen in Figure 8, it is preferable to conform the shapes of the pins to the shapes of the grooves in order to securely seat the assemblies in their proper band positions. For example, the engagement surfaces 1043a and 1043b of pin 1043 are preferably configured to seat securely against surfaces 134a and 134b of a groove on threaded rod 134. First, this configuration offers the advantage in that the pin (and consequently the assembly to which it is attached) is centered in the groove to obtain a reproducible and precise band position. Second, this configuration restricts movement of the pin in either direction along the longitudinal axis of threaded rod 134 since surfaces 1043a and 1043b are engaged with surfaces 134a and 134b, respectively, to restrict movement of the pin in the groove.

One skilled in the art will appreciate that other shapes of pins and grooves will provide the desired features similar to the shapes shown in Figure 8.

Returning to Figure 5, as the pins are engaging the grooves, some sliding motion will generally need to occur to allow the pins to fully engage the grooves.

Moreover, it is necessary that no significant residual forces reside within either of the assemblies 104' and 105' (e.g., forces which urge an assembly in either direction along the cross-scanning axis), inasmuch as such forces could cause the resultant positions to be less than totally reproducible. Some sticking of the bushings in each assembly may occur. To prevent this from happening, shaft 130 is made to rotate at the same time that the pins are engaging the grooves. This could be done by a separate stepping or other motor, but preferably is accomplished by gears 137 and 138, linking the shafts 130 and 132, both being linked to a single stepping motor 136.

The preferred pin and groove system disclosed herein offers a number of advantages over conventional pin and groove-type engagements which are typically used for positioning in systems which do not require the same degree of precision as is required in bandwise imaging.

For example, since the pin and groove are disengaged when the assemblies are translated, the pins and grooves of the present invention are not subjected to wear due to frictional engagement during translation, as is common with many conventional mechanical arrangements.

Further, the assemblies of the present invention are not subjected to forces along the cross-scanning axis when the wire rope translation system is not actuated, which allows the pins to seat securely in the center of the grooves. Many conventional systems utilize pin and groove systems to hold an assembly in tension against a groove, which does not allow precise seating in the groove, and which increases the wear on the pins and grooves.

Also, the pin and groove system of the present invention provides opposing contact surfaces for the pins in the grooves, which restricts motion in both directions along the cross-scanning axis when the pin is seated in the groove. This provides reproducible and precise positioning of the assemblies at each band position, and also secures the assembly from movement due to vibrations or other forces incident in a mechanical system. In contrast, many conventional pin and groove configurations only provide one contact surface to which a pin is biased against, essentially to selectively restrict motion in one direction. As discussed above, such a configuration does not provide as secure a seat for a pin, and it tends to induce wear on the pins and grooves.

Overlap Contour Control
Figure 9 shows the calculated integrated exposure of a portion of an image, for three bands side by side, for a constant grayshade, wherein there is no mismatch in the placement of bands. The ordinate in this graph is light intensity, and the abscissa is distance parallel to the drum's axis. Depending on the beam energy profile, the very fine variation in energy over the extent of a single pixel may be more or less than what is illustrated.

Figure 10 shows the same conditions, except that the central band of image is misplaced by 5 microns. In each of Figures 9, 10 and 11 it is assumed that there are 30 scan lines per band, at 300 dpi. As can be seen in Figure 10 there are significant discontinuities at the points 4 and 5 of joining of the bands. This is an undesirable situation as would result from the use of prior art technology.

In Figure 11, it is assumed that the number of beams per band is increased from 30 to 32, so that some overlap between bands can occur. Furthermore, the intensity of the overlapping beams is adjusted in accordance with the known amount of overlap. The data used to program the overlapping beams is the same, except multiplied by coefficients specific for given beams and given amounts of positional error. This adjustment is done by a recursive mathematical procedure which simply adjusts various coefficients until the variation from the ideal is minimized. The results of these calculations are placed in lookup tables. The intensity errors in Figure 10 are visible. Those in
Figure 11 are not visible. Figure 11 represents the result which would be obtained using the present invention.

It should be understood that when a band of an image is improperly placed, there are two consequences.

The first is that objects in different bands may be closer together or further apart than they should be.

Thin lines extending from one band to another will vary in thickness. A second consequence is that regions which are intended to have constant grayshades will have thin straight lines running through them at higher or lower grayshades. The human eye is much more adept at discerning the second consequence than it is at discerning the first one. The precision which is obtainable with the design disclosed herein or via various prior art methods is sufficient that distance errors, i.e. the first consequence, will not be visible.

The consequences of the second type are not eliminated via prior art methods, but are eliminated via the present invention.

The information which is used to select coefficients which determine the manner in which overlapping beams are programmed is based on the fixed band positioning errors for each position. These characteristics are determined by the optical scanning of a print produced with assumed zero error at each position. As a result of this scanned information, again by a recursive procedure, the necessary band positioning error information is developed.

In real time, the band positioning error information may be combined with drum runout information. Drum runout is the displacement of the drum parallel to its axis, resulting from debris in its bearings, vibration of the frame, and other factors.

For instance, in the embodiment of Figure 5, the optical interrupter 140, which senses a ring 141, is attached to the drum 101. The interrupter 140 is stationary, attached to the grooved part in such a way as to be mechanically stable despite internally or externally generated vibrations. The ring 141, in cross section, is "C" shaped. As the drum 101 moves, relative to the interrupter 140, a beam from a light emitting diode to a phototransistor (not shown), within the optical interrupter, is more or less interrupted. With appropriate circuitry, the degree of runout is measured.

A microprocessor system combines this information with the band positioning error information and other information to produce information which is used to control the way overlapping information is programmed.

Typically one side of each band is reduced in intensity at a fixed rate, and the other side is reduced in intensity at an adjustable rate. For example, the intensity multipliers of cells &num;0 and &num;31 would be nominally 0.333 and those for cells &num;1 and &num;30 would be .667. Cell &num;0 (33%) of one band would overlap cell &num;30 (67%) of the next band. Cell &num;1 (67%) of one band would overlap cell &num;31 (33%) of the next band. Cells &num;0 and &num;1 would be imaged at fixed intensity multipliers, while cells &num;30 and &num;31 would be adjusted, corresponding to errors in position and optical magnification ratio.

These errors are functions of both position (band &num;) and rotation angle (pixel &num;) of the drum. The drum can be expected to be not exactly circular, and not exactly centered about its shaft. Other reasons may also exist to cause the magnification ratio to vary (albeit reproducibly) from place to place. These conditions are reflected in the programming which the microprocessor system uses to calculate the effective extent of overlap between bands.

Lookup tables are used extensively to characterize and calibrate the intensities in the overlapping regions. This procedure is somewhat complex, and is detailed as follows.

First of all, an estimate of the beam energy profile is made. For example, if the light beams were shaped as a series of circles in cross-section which appeared in focus on the drum, then the relative beam energy y for a single beam would be a function of position x such that xA2 + y^2 = rA2, where r = the radius of the beam. On the other hand, if the beam were significantly defocused in the optical system, the beam energy profile would be approximately that of a gauss ivan curve. If the beam is in focus, but the shape is other than circular (for instance, if shaped by an-aperture), still another beam energy profile would be obtained.

Based on known factors, the best estimate of the beam energy profile is made.

That being done, a theoretical analysis is made to determine what coefficients should be used for various degrees of overlap. A person skilled in mathematics might work out an exact analysis. An easier approach is to utilize a computer and a trial and error method.

This may be done as follows. Six beams are considered:
Beam &num;1 at position 0 * s, intensity 100%
Beam &num;2 at position 1 * s, intensity 67%
Beam &num;3 at position 2 * s, intensity 33%
Beam &num;4 at position 1 * s + e, intensity cl
Beam &num;5 at position 2 * s + e, intensity c2
Beam &num;6 at position 3 * s + e, intensity 100% where s is the nominal beam spacing, and e is the positional error between one band and another. Ideally e = 0. If this were the case, cl = 33W-and c2 = 67% would produce the minimal (i.e. no) discontinuity between bands. The intention of the program is to calculate cl and c2 for various values of e.

An array of numbers is established in a program for positions including at least the range of 0 to 3 * s, at a resolution of .01 * r or finer. Over at least the region of overlapping bands, the quantity of the energy and the placement of each part of each beam is calculated, and the results added (integrated). Once the calculation is complete, the rms (root mean square) deviation from the ideal mean over the range of band overlap is calculated. Desirably, this rms error is minimized by selecting cl and c2.

Trial and error values of cl and c2 are used for each value of e. A systematic method is used to try various values of cl and c2, each slightly different from the old values. If the new values of cl and c2 are better (i.e. result in a lower rms error value) than the old ones, the new values are used. If not, the old values are used. This is a one-time calculation, so the fact that it is somewhat time-consuming for a computer to calculate is of no concern.

As an illustration, a circular beam profile is assumed, of radius 2.5 mils. A beam separation of 3.33 mils (300 dpi) is also assumed. A program as described above calculates the following values for cl and c2 as a function of error e, in mils:
e, mils cl c2
-2.00 0.17 0.46
-1.50 0.21 0.51
-1.00 0.24 0.56
-0.50 0.28 0.61
0.00 0.33 0.67
0.50 0.41 0.74
1.00 0.49 0.80
1.50 0.56 0.86
2.00 0.60 0.90
This information, that of relative intensity information for each cell, is stored within a lookup table. Part of the address information for the lookup table is the composite error information for each band.

Figure 12 shows in simplified block diagram form some of the circuitry which is used for controlling the individual PLZT cells. A DRAM, 150, is loaded with information pertaining to the desired graylevel for at least one color within one or more bands. This is done by circuitry which is not shown. The data from this memory is used as an address input to a second memory, lookup table memory 151. Other address inputs to this memory are the cell &num; and the processed runout extent (degree of overlap of images). The cell &num; has five possible values: 0, for cell 0; 1, for cell 1; 2, for cells 3 through 30; 3, for cell 31, and 4, for cell 32.

All of the central cells (cells 3 through 30) are treated identically. The runout information is presented as 7 bits of information. Thus the total number of address bits to this memory, 151, are 8 (for grayshade input) + 3 (cell &num;) + 7 (runout extent) = 18.

A 262K memory size is required for memory 151, which is easily accomplished.

As mentioned previously, the composite error information for each band is based on the sum of reproducible components (band position error and drum diameter variations) and random components (drum bearing runout) . The extent of the reproducible components is measured by scanning a photographic print prepared with assumed values of band position errors. The extent of the variable component is measured via the drum runout optical interrupter. This can be calibrated theoretically, by knowing the sizes of the light source and detector, or it can be calibrated by assuming various possible calibrations and determining which provide the best match to measured (scanned) photographic prints.

The output of memory 151 (8 bits of processed grayshade information), is fed as an address input to a second lookup table memory, 152. This memory contains nonlinearity information which characterizes and calibrates the individual PLZT chip cells, the drivers, the optical apertures, and all other relevant optical and mechanical components. Five additional address bits are used to indicate which of 32 PLZT cells is being used. The output of this memory drives one of 32 digital to analog converters. Each of these D/A converters drives a high voltage amplifier which, in turn, is connected to a corresponding PLZT cell.

The lookup table, 152, which contains the PLZT characteristics, is established by positioning the objective lens assembly so that a photodiode receives the light from the PLZT cells. One cell at a time is turned on, and the received light measured. For each cell, the results of drive signals from the minimum to the maximum are determined. By inspection of this data, it can be determined which drive voltage to use for any given desired optical output. This is done automatically at periodic intervals.

Not shown, but obvious to those skilled in the art, is various control electronics. A microprocessor system obtains the necessary information and loads this information into each of the memories.

A photodiode (not shown) may be positioned at the edge of the drum so that with proper location of the objective lens 1042 or 1042' this photoreceptor, rather than the drum, receives the light produced by the PLZT chip array. By programming the PLZT chip in various ways, certain characterization and calibration information is obtained. This data, together with data taken by scanning a print produced with assumed values of overlap for each band, provides the data used by the microprocessor system to adjust the image data in the overlapping region.

The above discussion, examples and embodiments illustrate my current understanding of the invention.

However, one skilled in the art will appreciate that various additional changes may be made without departing from the spirit and scope of the invention. Thus, the invention resides wholly in the claims hereafter appended.

Claims

I CLAIM:

1. An imaging apparatus for forming an image on photosensitive media comprising: (a) imaging means for applying a plurality of light beams to a photosensitive media; (b) first media transport means for moving the photosensitive media and the imaging means relative to one another in a first direction; (c) second media transport means for moving the photosensitive media and the imaging means relative to one another between a plurality of imaging bands defined along a second direction, the second direction being generally orthogonal to the first direction;

; (d) control means for, in a first period, operating the imaging means to apply the plurality of light beams to the photosensitive media when the photosensitive media and the imaging means are at one of the plurality of imaging bands such that an image band is exposed on the photosensitive media, and in a second period, actuating the second media transport means to move the photosensitive media and the imaging means relative to one another in the second direction to a relational position proximate one of the plurality of imaging bands; and (d) positioning means for positioning the photosensitive media and the imaging means relative to one another in the second direction at a discrete relational position corresponding to one of the plurality of imaging bands such that uniform spacing is provided between adjacent image bands exposed on the photosensitive media.

2. The imaging apparatus of claim 1, wherein the positioning means comprises positional sensing means for generating a positional signal representative of the position of the photosensitive media and the imaging means relative to one another in the second direction.

3. The imaging apparatus of claim 2, wherein the positional sensing means comprises a ruling and an optical interrupter, the ruling and optical interrupter being arranged to move relative to one another along a longitudinal axis of the ruling concurrent with the movement of the photosensitive media and the imaging means relative to one another in the second direction, wherein the ruling includes a plurality of alternating transparent and opaque portions disposed along the longitudinal axis of the ruling, and the optical interrupter is configured to sense the opacity of the ruling along the longitudinal axis of the ruling and provide the positional signal representative thereof.

4. The imaging apparatus of claim 3, wherein the ruling is fixably attached relative the photosensitive media in the second direction and the optical interrupter is operatively connected to an objective lens assembly on the imaging means which moves relative the photosensitive media in the second direction among the plurality of imaging bands, and wherein the ruling is configured such that movement of the optical interrupter a distance equal to a combined width of adjacent transparent and opaque portions results in a movement of the objective lens assembly a distance equal to a width of an imaging band, and such that when the objective lens assembly is disposed proximate a boundary between imaging bands, the optical interrupter is disposed proximate a boundary between adjacent transparent and opaque portions on the ruling.

5. The imaging apparatus of claim 4, wherein the second media transport means comprises a leadscrew, operatively connected to the imaging means, for moving the objective lens assembly relative the photosensitive media in the second direction among the plurality of imaging bands.

6. The imaging apparatus of claim 5, wherein the imaging means further comprises a mirror assembly configured to maintain a fixed optical distance between the objective lens assembly and a fixed portion of the imaging means, wherein the mirror assembly is operatively connected such that it is movable along the second direction by the leadscrew.

7. The imaging apparatus of claim 2, wherein the positioning means further comprises storage means for storing band positions representative of the discrete relational position for each of the plurality of imaging bands, wherein each band position is determined from the positional signal provided by the positional sensing means during characterization of the apparatus, and wherein when the control means moves the photosensitive media and the imaging means to one of the plurality of imaging bands, the positioning means actuates the second transport means to move the photosensitive media and the imaging means relative to one another until the positional signal from the positional sensing means is substantially equal to a reference positional signal representative of the band position stored in the storage means for the one of the plurality of imaging bands.

8. The imaging apparatus of claim 7, wherein the positioning means further comprises real-time characterization means for modifying the band positions stored in the storage means in response to the positional signals provided by the positional sensing means at each of the plurality of imaging bands.

9. The imaging apparatus of claim 8, wherein, at each of the plurality of imaging bands, the real-time characterization means compares the positional signal provided by the positional sensing means with an expected positional signal representative of the band position stored in the storage means for the imaging band, and modifies the band position stored in the storage means to compensate for deviations between the positional and expected positional signals.

10. The imaging apparatus of claim 7, wherein the positioning means further comprises fine-tuning means for actuating the imaging means to compensate for positional errors detected by the positional sensing means at each of the plurality of imaging bands.

11. The imaging apparatus of claim 10, wherein the fine-tuning means comprises means for actuating the imaging means to rotate a glass plate and deflect the plurality of light beams along the second direction in response to deviations between the positional signal provided by the positional sensing means and an expected positional signal representative of the band position stored in the storage means for the imaging band.

12. The imaging apparatus of claim 10, wherein the fine-tuning means comprises means for actuating the imaging means to control contour energies of the plurality of light beams in overlap regions between adjacent imaging bands in response to deviations between the positional signals provided by the positional sensing means and expected positional signals representative of the band positions stored in the storage means for the adjacent imaging bands.

13. The imaging apparatus of claim 1, wherein the imaging means comprises a plurality of PLZT light valves for controlling the plurality of light beams.

14. The imaging apparatus of claim 1, wherein the photosensitive media is disposed about a substantially cylindrical drum and wherein the first media transport means moves the photosensitive media and the imaging means relative to one another by rotating the drum.

15. The imaging apparatus of claim 1, wherein the plurality of light beams applied by the imaging means are arranged and configured so as to project light along an axis generally orthogonal to the first direction.

16. The imaging apparatus of claim 1, wherein the positioning means comprises a pin selectively engageable with a plurality of grooves disposed at a plurality of relative positions along a grooved member, wherein a longitudinal axis of the grooved member is defined along the second direction.

17. The imaging apparatus of claim 16, wherein the plurality of grooves are fixably attached relative the photosensitive media in the second direction and the pin is operatively connected to an objective lens assembly on the imaging means which moves relative the photosensitive media in the second direction among the plurality of imaging bands, and wherein each of the plurality of grooves is configured such that, when the pin is selectively engaged with the groove, the groove restricts movement of the pin in both directions along the longitudinal axis of the grooved member.

18. The imaging apparatus of claim 17, wherein the positioning means comprises means for disengaging the pin from one of the plurality of grooves when the control means actuates the second media transport means and means for engaging the pin with one of the plurality of grooves when the control means has moved the imaging means to a relational position proximate one of the plurality of imaging bands.

19. The imaging apparatus of claim 18, wherein the objective lens assembly is mounted to a rotatable shaft, and wherein the positioning means further comprises means for rotating the shaft concurrently with engaging the pin with one of the plurality of grooves.

20. The imaging apparatus of claim 19, wherein the second media transport means is configured such that the objective lens assembly is unbiased in both directions along the longitudinal axis of the grooved member when the pin is engaged with one of the plurality of grooves.

21. The imaging apparatus of claim 1, wherein the plurality of light beams includes first and second light beams oriented such that a scan line imaged by the first light beam in one image band is overlapping with a scan line imaged by the second light beam in an adjacent image band, wherein the imaging means reduces the intensity of the first and second light beams to provide a combined intensity for the overlapping scan lines, and wherein the positioning means further comprises means for adjusting the combined intensity of the first and second light beams to compensate for band positioning errors.

22. The imaging apparatus of claim 1, wherein the positioning means further comprises means for compensating for positional error of the photosensitive media along the second direction.

23. A spatial stabilization apparatus for use in a bandwise imaging system of the type having an imaging device, a portion of which is configured to apply a band of light beams to a photosensitive media as the photosensitive media moves relative to the portion of the imaging system in a scanning direction, and a media transport mechanism for moving the photosensitive media and the portion of the imaging device relative to one another between a plurality of imaging bands in a crossscanning direction which is generally orthogonal to the scanning direction, the apparatus comprising a band positioning controller, operable during a dead cycle in which scanning is disabled on the bandwise imaging system,

the band positioning controller being configured to actuate the media transport mechanism to move the photosensitive media and the portion of the imaging device relative to one another in the cross-scanning direction to a relational position proximate one of the plurality of imaging bands, and to position the photosensitive media and the portion of the imaging device relative to one another in the cross-scanning direction at a discrete relational position corresponding to one of the plurality of imaging bands such that uniform spacing is provided between adjacent imaging bands exposed on the photosensitive media.

24. The spatial stabilization apparatus of claim 23, further comprising: (a) a positional sensor for generating a positional signal representative of the position of the photosensitive media and the portion of the imaging device relative to one another in the cross-scanning direction; and (b) a storage device for storing a band position representative of a discrete relational position for each of the plurality of imaging bands such that the band positioning controller actuates the media transport mechanism to move the photosensitive media and the portion of the imaging device relative to one another to the discrete relational position for one of the plurality of imaging bands, wherein each band position is determined from the positional signal provided by the positional sensor during characterization of the bandwise imaging system.

25. The spatial stabilization apparatus of claim 24, wherein the band positioning controller includes a real-time characterization routine for modifying the band positions stored in the storage device in response to the positional signals provided by the positional sensor at each of the plurality of imaging bands.

26. The spatial stabilization apparatus of claim 24, wherein the band positioning controller actuates the imaging device to rotate a glass plate disposed in an optical path of the band of light beams and deflect the band of light beams in the cross-scanning direction in response to deviations between the positional signal provided by the positional sensor and an expected positional signal representative of the band position stored in the storage device for the imaging band.

27. The spatial stabilization apparatus of claim 24, wherein the band positioning controller actuates the imaging device to control contour energies for the band of light beams in overlap regions between adjacent imaging bands in response to deviations between the positional signals provided by the positional sensor means and expected positional signals representative of the band positions stored in the storage device for the adjacent imaging bands.

28. A method for spatially stabilizing in a bandwise imaging system of the type having scanning means for applying a band of light beams to a photosensitive media as the photosensitive media moves relative to the scanning means in a scanning direction, and media transport means for moving the photosensitive media and the scanning means relative to one another between a plurality of imaging bands in a cross-scanning direction which is generally orthogonal to the scanning direction, the method comprising the steps of: (a) actuating the media transport means, during a dead cycle in which scanning is disabled on the bandwise imaging system, to move the photosensitive media and the scanning means relative to one another in the cross-scanning direction to a relational position proximate one of the plurality of imaging bands;

and (b) positioning the photosensitive media and the scanning means relative to one another in the cross-scanning direction at a discrete relational position corresponding to one of the plurality of imaging bands such that uniform spacing is provided between adjacent image bands exposed on the photosensitive media.

29. The method of claim 28, wherein the positioning step comprises the steps of: (a) sensing the position of the photosensitive media and the scanning means

relative to one another in the cross-scanning direction and generating a positional signal representative thereof; (b) retrieving a band position from a storage media having band positions representative of discrete relational positions for each of the plurality of imaging bands, wherein the band positions are determined from positional signals generated during characterization of the bandwise imaging system; and (c) actuating the media transport means to position the photosensitive media and the scanning means relative to one another such that the positional signal is substantially equal to a reference positional signal representative of the band position retrieved from the storage media.

30. The method of claim 29, wherein the positioning step further comprises the step of modifying the band positions stored in the storage media in response to the positional signals generated at each of the plurality of imaging bands.

31. The method of claim 30, wherein the positioning step further comprises the step of rotating a glass plate disposed in an optical path of the band of light beams to deflect the band of light beams in response to deviations between the generated positional signal and an expected positional signal representative of the band position stored in the storage means for the imaging band.

32. The method of claim 30, wherein the positioning step further comprises controlling the contour energies of the band of light beams in overlap regions between adjacent imaging bands in response to deviations between the generated positional signals and expected positional signals representative of the band positions stored in the storage media for the adjacent imaging bands.