JP2006229264A - Method and system for adaptive dither structure - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve such a problem that there remains a segment of a grayscale where it is difficult to remove contour artifacts without increasing the amplitude of a dither pattern (e.g., noise) so much that it becomes visible. <P>SOLUTION: To resolve the above problem, a dither array in which both the spatial dimensions and the temporal dimensions have high-pass characteristic is combined with a color image. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to dither technology, and more particularly to a method and system for an adaptive dither structure.

  A digital image is transmitted with values that indicate the luminance and color attributes of the image in an array of locations within the entire image. Each value is represented by a predetermined number of bits. When bandwidth, storage and display requirements are not limited, enough bits are available to display the image with visual clarity and practical color reproduction that are not virtually prohibited. However, when the bit depth is limited, adjacent luminances, i.e. gradations between color levels, become perceptible and can even be uncomfortable for a human observer. Such an effect is evident as a contouring artifact that can be seen in images with low bit depth. The contour lines become apparent in the low frequency region where the luminance changes slowly, in which case the pixel values are moved to one side or the other side of the coarse gradation step.

  These contouring artifacts can generally be “decomposed” by adding noise or other dither patterns to the image before reducing quantization or other bit depth. This addition of noise or pattern results in random, pseudo-random, or other changes in pixel values, which reduces the occurrence of contours and their visibility. In general, images are perceived as more natural and comfortable for human observers.

  Some of these methods may be described with reference to FIG. In these systems, noise or dither pattern 16 can be added (4) to image 2 or otherwise combined. This combined image is then quantized (6) to a low bit depth. This image may then be displayed directly or transmitted (8) to the receiver 10 as shown in FIG. After reception, the noise / dither 16 added to the image can be subtracted (12) from the image, or otherwise reverse-combined, to reduce the noise / dither visual effects in areas where contouring is less likely to occur. This image is then displayed (14) on the receiver side. These methods can also be used in systems that do not transmit and receive, such as displays that have a lower bit depth capability than the image data 2 to be displayed.

  Some of these methods can be described with reference to FIG. In these systems 20, image 2 is combined (28) with noise / dither pattern 16 and sent to a display system 22 that cannot display the full range of image data contained within the image. These display systems 22 can quantize (24) the image data to a bit depth that matches the display capabilities. Next, the quantized image data is displayed on the display 26.

  In the system shown in FIG. 2, the noise / dither pattern is not subtracted or reversed from the image. In these systems, less noise can be added to the image before it causes undesirable visual effects, or “particleization”. It has been found that various frequency distributions for noise / dither patterns are more or less visible to the human visual system. In general, the human visual system functions as a low-pass filter that removes high-frequency data with a filter. Therefore, noise concentrated on the high frequency range is less visible than low frequency noise.

  Often, dither / noise patterns that are as large as the image file cannot be used. In these cases, smaller dither patterns can be used by repeating the pattern in rows and columns throughout the image. This process is often referred to as tiling. In multiple image sets, such as frames or fields of video images, the dither pattern can be repeated from frame to frame. This dither pattern can be designed to minimize artifacts caused by repetitive patterns.

The present invention is constituted by the following technical means in order to solve the above-described problems.
The first technical means of the present invention includes a first upper horizontal spatial frequency boundary, a first lower horizontal spatial frequency boundary, a first upper vertical spatial frequency boundary, a first lower vertical spatial frequency boundary, and a first upper Setting a first multidimensional array of dither pattern tiles including a temporal frequency boundary and a first lower temporal frequency boundary, wherein the spatial frequency boundary and the temporal frequency boundary constitute a high pass pattern configuration. , Second upper horizontal spatial frequency boundary, second lower horizontal spatial frequency boundary, second upper vertical spatial frequency boundary, second lower vertical spatial frequency boundary, second upper time frequency boundary and second lower time Setting a second multi-dimensional array of dither pattern tiles including frequency boundaries, wherein the second lower time frequency boundary is the first lower time frequency Associating the first multi-dimensional array of dither pattern tiles with a first range of image characteristic values lower than the field, and the second multi-dimensional array of dither pattern tiles and a second of image characteristic values. And a step of associating the range with the range.

  According to a second technical means, in the first technical means, the image characteristic value includes a luminance value.

  According to a third technical means, in the first technical means, the image characteristic value includes a local average luminance value.

  A fourth technical means includes receiving a digital image, designating a location in the image to apply a dither pattern tile, determining a local image characteristic for the tile location, Selecting a first dither pattern tile set from the dither pattern tile set, wherein the selection is based on the local image characteristics.

  A fifth technical means divides the digital image into tile block locations, determines a local luminance characteristic for each of the tile block locations, and selects one dither pattern array from the plurality of dither pattern arrays. And each of the arrays is associated with a range of local brightness characteristic values.

  The sixth technical means includes a first upper horizontal spatial frequency boundary, a first lower horizontal spatial frequency boundary, a first upper vertical spatial frequency boundary, a first lower vertical spatial frequency boundary, and a first upper time frequency boundary. And setting a first multidimensional array of dither pattern tiles, including a first lower temporal spatial frequency boundary, a second upper horizontal spatial frequency boundary, a second upper horizontal spatial frequency boundary, and a second upper Including a vertical spatial frequency boundary, a second lower vertical spatial frequency boundary, a second upper time frequency boundary, and a second lower time frequency boundary, wherein the second lower time frequency boundary is the first lower time frequency boundary Setting a second multi-dimensional array of dither pattern tiles that is lower than the first multi-dimensional array of dither pattern tiles and a first level of local luminance values. Associating the second multi-dimensional array of dither pattern tiles with a second range of local luminance values, dividing the digital image into tile block locations, Determining a local luminance characteristic for the tile block location and applying to the first set of tile block locations when the first tile block location has a local luminance characteristic value falling within the first range. Selecting a dither pattern tile from the first multi-dimensional array of dither pattern tiles for the tile block location when the second tile block location has a local luminance characteristic value falling within the second range Of dither pattern tiles for application to the second set of A step of selecting a dither pattern tile from serial second multidimensional array is obtained by comprising the.

  The seventh technical means includes a first upper horizontal spatial frequency boundary, a first lower horizontal spatial frequency boundary, a first upper vertical spatial frequency boundary, a first lower vertical spatial frequency boundary, and a first upper time frequency boundary. And a first multi-dimensional array of dither pattern tiles, wherein the first multi-dimensional array is associated with a first range of local luminance characteristic values, and a second upper A horizontal spatial frequency boundary, a second lower horizontal spatial frequency boundary, a second upper vertical spatial frequency boundary, a second lower vertical spatial frequency boundary, a second upper time frequency boundary, and a second lower time frequency boundary, A second dither pattern tile in which the second lower time frequency boundary is lower than the first lower time frequency boundary and the second multidimensional array is associated with a second range of local luminance characteristic values. Many The original array, based on local luminance characteristic value for the location in the image, is obtained by comprising the, a selector for selecting either of said first array and said second array.

  The eighth technical means includes an image storage device for storing an image to be displayed, a first upper horizontal spatial frequency boundary, a first lower horizontal spatial frequency boundary, a first upper vertical spatial frequency boundary, a first A lower vertical spatial frequency boundary, a first upper temporal frequency boundary, and a first lower temporal frequency boundary, wherein the first multidimensional array is associated with a first range of local luminance characteristic values A first multidimensional array of pattern tiles, a second upper horizontal spatial frequency boundary, a second lower horizontal spatial frequency boundary, a second upper vertical spatial frequency boundary, a second lower vertical spatial frequency boundary, a second Including an upper time frequency boundary and a second lower time frequency boundary, wherein the second lower time frequency boundary is lower than the first lower time frequency boundary, and the second multidimensional array has a local luminance characteristic value of Second len A dither array storage having content with a second multidimensional array of dither pattern tiles associated with the image, a processor for determining a local luminance characteristic value for the image location, and a local for the image location And a selector for selecting one of the first array and the second array based on a luminance characteristic value.

  The ninth technical means is characterized in that the image signal is combined with a dither array having a structure in which a spatial dimension and a temporal dimension have high-pass characteristics, and the image signal combined with the dither array is quantized and displayed. It is what.

  The dither structure can include multiple dither patterns to be used across a single multi-frame image. A three-dimensional dither structure as shown in FIG. 3 can utilize a series of dither patterns. These patterns 30 to 36 can be arranged in a sequence used in a sequential frame of video. The first dither pattern tile 30 can be used in the first video frame 38, while the next sequential pattern 32 can be used in the subsequent video frame 40. After using each pattern in the sequence, the sequence of patterns 30-36 can be repeated. These sequences can also be designed to particularly reduce the occurrence of artifacts resulting from repeated time patterns.

Next, the drawings will be described in detail. In the drawings, like numerals indicate like parts of the invention.
The methods and systems of embodiments of the present invention relate to display algorithms, processes and apparatus that use spatio-temporal dithering to increase the perceived bit depth. These methods and systems can be used for LCD or similar displays with digital bit depth bottlenecks, such as graphic controller cards with limited video RAM (VRAM). The LCD display itself or its internal hardware driver also has bit depth limitations. In embodiments of the present invention, the time response characteristics of the display can be used to dynamically parameterize the dither pattern.

  The overall problem is that image quality is lost so that the gray level per color is too low. This manifests itself as contouring and loss of information (especially loss of low amplitude signals). Embodiments of the present invention can be applied so that a 4-6 bit / color display can visually display an image having an image quality equivalent to 8 bits / color. Another application is to allow an 8-bit display to have 10-bit image quality when an image of 10 bits or more is to be displayed.

  A major problem at the current technical level is that grayscale segments remain in contour artifacts that are difficult to remove unless the amplitude of the dither pattern (eg, noise) is visibly increased. This is because the nonlinearity of the visual system is close to 1/3, while the tone scale shape is close to 2.4 to the gamma power. As a result of these two nonlinear cascade effects, the slope is steeper in darker areas than the rest of the tone scale.

  With reference to FIG. 4, some attributes of the embodiment of the present invention will be described. FIG. 4 shows a result of optimizing the pattern to the human visual system (HVS) characteristics (eg by defining the shape of the spatio-temporal color power spectrum to match the power spectrum of the equivalent noise of the visual system). 1 illustrates a method and system for generating a dither pattern. In these embodiments, the color image having the multi-color channels 52-56 and the dither array are combined before performing the quantization 62-66. This dither array structure is created with reference to a human visual system 68 that is not sensitive to high frequency noise. Accordingly, a space-time high-pass dither structure 70 is created in which both the spatial dimension and the temporal dimension have high-pass characteristics.

  In embodiments of the present invention, the dither structure can be optimized for display characteristics. In some of these embodiments, the display's inherent fixed pattern spatial noise can be used as an element in dither pattern design.

  The purpose for obtaining an effective dither pattern design is to add as much noise as possible and to ensure that no noise is seen in the displayed image. Embodiments of the present invention can utilize the LPF characteristics of the visual system so that the dither structure can be given high-pass characteristics and thus the dither pattern on the display can be attenuated by the LPF of the visual system mainly due to optical characteristics. In other words, the frequency response of the visual system (which is often modeled as the reciprocal of the contrast sensitivity function (CSF) similar to the frequency response) is used to obtain a dither pattern (noise ) Can be defined.

  Embodiments of the present invention can be used with displays that can display signals that change over time. In these embodiments, it is worth using a space-time dither structure. FIG. 5 shows the equivalent noise and the resulting dither pattern in icon form. Their power spectra are mutually high-pass in spatial and temporal frequencies. The Nyquist limit for digital frequency is 0.5 (0.5 cycle / sample; sample = pixel or frame). The axes include horizontal spatial frequency (H SF) 80, vertical spatial frequency (V SF) 82, and temporal frequency (TF) 84.

  In these space-time embodiments, the dither array can be stored in a series of 2D tiles (ie, equivalently as a 3D sequence), where the series consists of different sequential tiles for the sequential frame of the real-time display. FIG. 6 shows the behavior of the frame-synchronized tile selector 72 as shown in FIG. FIG. 6 shows the space-time dither structure imposed on sequential image frames, namely frame “p” 90 and frame “p + 1”. Dithered tiles are imposed on the image frame. There are a series of dither tiles that are added to the input image frame at any spatial location on the display, and these tiles are stepped sequentially (ie, FIG. 5) to preserve the temporal power spectrum of these designs.

  Some embodiments of the present invention can use a tile stepping method as shown in FIG. 6 to further reduce the likelihood of visual artifacts. In these embodiments, a space-time array of dither pattern tiles 110 may be used. These dither pattern tiles 110 are generally smaller than images that use these tiles to reduce memory size. Smaller tiles can cover images in tile patterns that use the same tile repeatedly. In some applications, the same tile can be used repeatedly across an image, but using this method can result in visual artifacts due to repeated patterns. This problem can be reduced or eliminated by using tiles from multiple consecutive frames. This method can be used in the spatial and temporal dimensions.

  As shown in FIG. 6, start from the first tile frame 94 and then traverse the image frame 90 to fill the tile pattern across the image 90 using each successive tile frame 96, 98 and 100. You can increment tiles spatially. This pattern of continuous tile frames can also be used in the time direction. The next successive image frame 92 uses the tile frame that follows the tile frame used in the previous image frame at a given tile location. For example, when using the first tile frame 94 in the upper left position within the first image frame 90, the next subsequent tile frame 96 is used at that location in the next image frame 92. Similarly, the second tile position in the first frame 90 is occupied by the second tile frame 96, and its position in the second image frame 92 is occupied by the third tile frame 98. The same pattern is repeated for each tile location and each image frame. Once the number of tile frames is exhausted, the order of tile sets can be repeated.

  In some applications, it can prove difficult to use the spatial characteristics of the display (except for straight forward use of PPI resolution and view distance in CSF mapping to the digital frequency domain). This is because the use of spatial display noise requires high resolution 2D imaging of the display, and the spatial modulation transfer function (MTF) is better than the eye limit, so the use of MTF has a significant impact. This is due to not giving. Thus, some applications only use the limits of the visual system spatially.

However, the temporal characteristics of the display cannot tune the dither array in that dimension. FIG. 7 shows various records consisting of temporal edge transitions for a particular “normal white” mode LCD. The vertical axis 120 is the luminance in cd / m ² units, and the horizontal axis 122 is in ms units. Since the stimulus is a square wave with respect to time, both the light and dark portions 124 and the light and dark portions 126 having different amplitudes can be seen. Note how fast the response is in the dark region as well as in the dark region.

  By measuring the time taken for the luminance change to go from 10% to 90%, each response is generally summarized by a single number. FIGS. 8 and 9 show the different amplitudes of the dark-to-light transition and the response from the light-to-dark transition to two key type LCDs (usually white and usually black), respectively. It is shown.

  Normally, the black mode has a slow response in the dark region of the tone scale, and the difficult region is in this region. Therefore, these slow responses can be used advantageously.

  Some embodiments of the present invention use a space-time dither pattern having a mutually high-pass spatial spectrum and a high-pass temporal spectrum where the low frequency cutoff varies with gray level. FIG. 10 shows this spectrum, which can be compared with another example (FIG. 5) having a fixed cutoff. Of course, the blocks shown in FIG. 10 are icons and the noise is not restricted to have a sharp cutoff frequency, and the noise is well visualized as a Gaussian block centered around these high frequencies. The key aspect of these embodiments is that the low time frequency cut-off is variable (indicated by a dotted line).

  In some embodiments, the variance value increases as the volume of the icon cube increases. Increasing the variance value can reduce the contour more greatly, which in turn can reduce the bit depth and more completely remove the contour artifacts in the troubled areas of the tone scale.

  FIG. 11 is a block diagram illustrating some embodiments of the present invention. In these embodiments, multi-dither pattern structures, ie, arrays 164, 166, and 168, are created and / or used. Dither pattern structures 164, 166 and 168 are created and stored before using the dither pattern tiles. Creation of the pattern starts with dividing the luminance spectrum into finite ranges 170, 172 and 174. For each of these ranges 170, 172 and 174, a different set or array of dither patterns is designed and generated (176, 178 and 180). These sets or arrays of dither patterns may vary not only in time bandwidth or lower time frequency cutoff, but also in other characteristics. These dither pattern sets or arrays can be generated by filtering noise into pattern specifications, dynamically creating patterns, or otherwise. Once the pattern set or array is generated, it can be stored for application to the image.

  In some embodiments, a set or array of dither patterns can be stored (164, 166 and 168) in the display device for internal use.

  These dither pattern sets can be applied to monochrome images and color images. In the color image embodiment, one image can be segmented according to the color channels 142, 144, and 146. In the embodiment shown in FIG. 11, the color channels correspond to the red, green and blue channels of the RGB display, but other color combinations can be used.

  Each color channel image frame 142, 144, and 146 and dither pattern tiles are combined before quantization. However, the particular dither pattern tile set selected for the tile location in the frame depends on the brightness level in the image frame to which this dither pattern tile is applied. For example, when the luminance level at a specific tile location falls within the first category or range 170, a dither pattern set 168 suitable for this range is selected by the tile selector 160 and applied. If the luminance value at the second location falls within the second category or range 172, another dither pattern set 166 can be selected by the tile selector 160.

  In some of these embodiments, a series of dither array sequences 164, 166 and 168 can be stored in memory within the display, and these sequences can be switched or selected based on the average luminance gray level of the image corresponding to the position of the tile. can do. The brightness level for a particular location in the image can be determined in a number of ways. The average luminance gray level of the tile area can be used, but other luminance data can be used in both dither pattern set design and dither pattern set selection during application of the dither pattern set. To prevent boundary effects associated with switching from one set of dither patterns to another, a transient region can be utilized to mix the two sets of dither patterns. For example, if the transition level between dither pattern set 1 and dither pattern set 2 is at average brightness level 64, instead of switching from set 1 to set 2 at 64, start at 60 and end at 68 The contribution of set 2 is gradually mixed into set 1.

  Once the dither pattern set is applied to the image, each color channel is quantized (152, 154 and 156). Further processing can also be performed. Finally, the quantized information is assigned to the display element and displayed to the user 158.

Dither Spectrum Generation Embodiments of the present invention include methods and systems for generating a dither spectrum. These arrays, sets or structures of dither patterns can be generated in several ways. In some embodiments, the white space-time spectrum (i.e. white to space and time Nyquist frequencies) can be filtered to generate the appropriate set of structures. In another embodiment, a set of dither patterns can be generated by filling the array using negative space-time color feedback.

  In some embodiments using spatial-temporal white spectral filtering, the starting point can be a 3D image array, where the dimensions of this array are horizontal space (pixels) filled with white spectrum, vertical space ( Pixel) and time (frame). In some of these examples, the reciprocal of the spatial CSF of the visual system (ie, S. Daily (1993) 17th in "Digital Image and Human Vision" edited by AB Watson, published by MIT Press). A filter that approximates the first spatially filtered noise within each frame can be generated by a filter that approximates the chapter (converted to low-pass form as described in the chapter (incorporated herein by reference)). . The result is then filtered over time by the reciprocal of the product of LCD time MTF and visual time CSF. The temporal MTF of an LCD can be non-linear overall, but is nearly linear for small amplitudes, and its shape varies as a function of gray level (as shown in the diagonal areas of FIGS. 5 and 6). The LCD time MTF can be calculated from the edge response using a normal line spreading function (LSF) calculation. An approximation can be used for each of these filters, and the Gaussian filter is a good first order approximation.

  An array of dither patterns can also be generated by filling the array with negative space-time color feedback. In some embodiments, a repulsion function can be used to sequentially assign dither values to locations that result from a desired pattern. Based on the size of the dither array, each gray level occurs a fixed number of times in the tile, which results in a uniform pdf as desired. A possible position for each gray level is then assigned based on the visibility of the array obtained by using the visual error function. The visual error function is based on a spatio-temporal CSF model, generally MSE weighted to CSF.

  Embodiments of the present invention include monochrome and color methods and systems. In color applications, several dither pattern arrays can be generated using three independent spatio-temporal arrays whose luminance is decorrelated across the array. This is an attempt to make the RGB array have equal brightness.

  Another embodiment of the invention includes a dither pattern generated in real time. In some of these embodiments, the local gray level parameter can control the dither generation process. In these embodiments, the time bandwidth can be changed with respect to the gray level parameter. Thus, in some examples, it can be seen that the lower boundary of the time bandwidth and the variance value change.

  Embodiments of the invention can include any number of dither pattern sets and any number of gray level ranges corresponding to these sets. In a simple embodiment, only two space-time noise sets are used. One set is used for the brighter range of gray levels and the other set is used for the darker range. The set used for the dark range has a narrow time bandwidth and a large variance value.

  In some embodiments, a color array can be generated by starting with a number of separate arrays. These arrays are then applied to the mating color signal and from a non-color signal and two color signals (eg L *, A * and B *, or Y, U and V) through the matrix to a 3-channel RGB signal. Converted.

  The foregoing detailed description has set forth a number of specific details to provide a thorough understanding of the present invention, but those skilled in the art will be able to practice the invention without departing from these specific details. You can understand what you can do. In other instances, well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the present invention.

  All references cited herein are incorporated by reference.

  The terms and expressions used in the above description are terms used in the description, not as terms that limit the invention, and in using such terms and expressions, It will be appreciated that some equivalents thereof are not excluded and that the scope of the invention is defined solely by the appended claims.

It is a figure which shows the conventional display device. It is a figure which shows another conventional display device. FIG. 6 illustrates a conventional dither structure as applied to an image frame. 1 is a diagram showing a space-time dither system. FIG. It is a figure which shows the icon display of a mutually high pass space and a high pass time dither spectrum. FIG. 6 is a diagram illustrating a dither structure having tiles applied to an image frame. Figure 6 is a graph showing the time response of an LCD display to different gray level transients. It is a graph which shows the transition time with respect to always white LCD. It is a graph which shows the transition time with respect to always black LCD. FIG. 6 is an icon display showing a high-pass space and a high-pass time dither spectrum with respect to each other where the lower boundary of the time frequency range is variable. FIG. 2 illustrates a system that uses a dither tile set that relies on multiple gray levels.

Explanation of symbols

52, 54, 56 ... color channel, 58 ... RGB tile, 62, 64, 66 ... quantizer, 68 ... visual system, 70 ... visual noise model, 72 ... tile selector.

Claims (9)

a. First upper horizontal spatial frequency boundary, first lower horizontal spatial frequency boundary, first upper vertical spatial frequency boundary, first lower vertical spatial frequency boundary, first upper time frequency boundary, and first lower time frequency Setting up a first multidimensional array of dither pattern tiles including boundaries,
b. The spatial frequency boundary and the temporal frequency boundary constitute a high-pass pattern configuration,
c. Second upper horizontal spatial frequency boundary, second lower horizontal spatial frequency boundary, second upper vertical spatial frequency boundary, second lower vertical spatial frequency boundary, second upper time frequency boundary, and second lower time frequency Setting a second multidimensional array of dither pattern tiles including boundaries,
d. The second lower time frequency boundary is lower than the first lower time frequency boundary;
e. Associating the first multi-dimensional array of dither pattern tiles with a first range of image characteristic values;
f. Associating the second multidimensional array of dither pattern tiles with a second range of image characteristic values. A method for creating a dither pattern structure.   The method of claim 1, wherein the image characteristic value includes a luminance value.   The method of claim 1, wherein the image characteristic value comprises a local average luminance value. a. Receiving a digital image;
b. Designating a location in the image to apply a dither pattern tile;
c. Determining local image characteristics for the tile location;
d. Selecting a first dither pattern tile set from a plurality of dither pattern tile sets, wherein the selection is based on the local image characteristics to adaptively process a digital image the method of. a. Dividing the digital image into tile block locations;
b. Determining a local luminance characteristic for each of the tile block locations;
c. Selecting a dither pattern array from a plurality of dither pattern arrays, wherein each of the arrays is adaptively adapted to a digital image characterized by being associated with a range of the local luminance characteristic values Way to handle. a. First upper horizontal spatial frequency boundary, first lower horizontal spatial frequency boundary, first upper vertical spatial frequency boundary, first lower vertical spatial frequency boundary, first upper time frequency boundary, and first lower time frequency Setting up a first multidimensional array of dither pattern tiles including boundaries,
b. Second upper horizontal spatial frequency boundary, second lower horizontal spatial frequency boundary, second upper vertical spatial frequency boundary, second lower vertical spatial frequency boundary, second upper time frequency boundary, and second lower time frequency Setting a second multidimensional array of dither pattern tiles that includes a boundary, wherein the second lower time frequency boundary is lower than the first lower time frequency boundary;
c. Associating the first multidimensional array of dither pattern tiles with a first range of local luminance values;
d. Associating the second multidimensional array of dither pattern tiles with a second range of local luminance values;
e. Dividing the digital image into tile block locations;
f. Determining local luminance characteristics for a number of said tile block locations;
g. Dither from the first multi-dimensional array of dither pattern tiles for application to a first set of tile block locations when the first tile block location has a local luminance characteristic value that falls within the first range. Selecting a pattern tile;
h. Dither from the second multi-dimensional array of dither pattern tiles for application to a second set of tile block locations when the second tile block location has a local luminance characteristic value that falls within the second range. Selecting a pattern tile, and adaptively processing the digital image. a. First upper horizontal spatial frequency boundary, first lower horizontal spatial frequency boundary, first upper vertical spatial frequency boundary, first lower vertical spatial frequency boundary, first upper time frequency boundary, and first lower time frequency A first multidimensional array of dither pattern tiles including a boundary, wherein the first multidimensional array is associated with a first range of local luminance characteristic values;
b. Second upper horizontal spatial frequency boundary, second lower horizontal spatial frequency boundary, second upper vertical spatial frequency boundary, second lower vertical spatial frequency boundary, second upper time frequency boundary, and second lower time frequency A dither pattern including a boundary, wherein the second lower time frequency boundary is lower than the first lower time frequency boundary, and wherein the second multidimensional array is associated with a second range of local luminance characteristic values A second multidimensional array of tiles;
c. A digital image processing apparatus, comprising: a selector for selecting one of the first array and the second array based on a local luminance characteristic value for a location in the image; System for. a. An image storage device for storing an image to be displayed;
b. i. First upper horizontal spatial frequency boundary, first lower horizontal spatial frequency boundary, first upper vertical spatial frequency boundary, first lower vertical spatial frequency boundary, first upper time frequency boundary, and first lower time frequency A first multidimensional array of dither pattern tiles including a boundary, wherein the first multidimensional array is associated with a first range of local luminance characteristic values;
ii. Second upper horizontal spatial frequency boundary, second lower horizontal spatial frequency boundary, second upper vertical spatial frequency boundary, second lower vertical spatial frequency boundary, second upper time frequency boundary, and second lower time frequency A dither pattern including a boundary, wherein the second lower time frequency boundary is lower than the first lower time frequency boundary, and wherein the second multidimensional array is associated with a second range of local luminance characteristic values A dither array storage device having content with a second multidimensional array of tiles;
c. A processor for determining a local luminance characteristic value for the image location;
d. An image display device comprising: a selector for selecting one of the first array and the second array based on a local luminance characteristic value for the image location.   An image display device characterized by combining an image signal with a dither array having a structure in which a spatial dimension and a temporal dimension have high-pass characteristics, and quantizing and displaying the image signal combined with the dither array.

Images