Patent Application
20070070427
The present invention relates generally to systems and methods for enhancing printability of documents, and more particularly, to systems and methods that enhance printability of document images based upon the content of such documents.
Error diffusion is one of the most pervasive techniques for generating halftones (as for printing) from contone raster images. Error diffusion avoids aliasing artifacts to which thresholding is vulnerable, but primitive error diffusion implementations are not well-suited to print technologies such as electrophotographic or “laser” printing with the strong interdependencies among adjacent dots.
One set of error diffusion techniques for clustering dots involves hysteresis in the diffusion. Spatial hysteresis requires more extensive error buffering, which represents an appreciable cost for high-resolution implementations. Alternatively, clustering may be forced by adding spatially-dependent dither to error diffusion's threshold. However, visible artifacts result from interference between the frequencies of spatially dithered thresholds and error diffusion's natural, tone-dependent frequencies. This interference can be avoided by applying a lower amplitude clustering dither to input in lieu of (or in coordination with a lesser amount of) threshold dither. This approach for clustering dots is described further in the U.S. Non-Provisional Patent Application titled “Apparatus and Method for Error Diffusion with Dither”, filed on Aug. 18, 2005, and assigned to the assignee of the present invention, the contents of which are incorporated herein as if set forth fully herein.
Nevertheless, techniques to interfere with error diffusion artifacts or cluster output also disrupt the reproduction of spatial details from contone raster images. Therefore, what is needed is error diffusion that applies the results of pixel classification to adjust the behavior of the error diffusion.
According to the present invention, pixel classification guides the manner in which error diffusion is performed. For inputs with slowly changing values (i.e., low spatial frequencies), clustered error diffusion may be used to improve uniformity. In such a scenario degradation of detail is not relevant because detail is also absent in the input. In contrast, for input with rapidly changing values, (i.e., high spatial frequencies), the present invention does not apply spatial dither. In those situations uniformity is not relevant since the input itself is highly varying.
According to an embodiment of the present invention, pixel classification determines whether spatially-dependent clustering dither is utilized. Pixels classified as low-frequency (i.e., pixels in a smooth area) use clustering dither while high-frequency pixels (e.g., edges) do not. According to an aspect of the present invention, a range of strengths of the clustering dither may also be selected based upon the relative frequency in a pixel's neighborhood. The strength of the clustering dither may be greatest with the lowest-frequency input and least with the highest-frequency input, with intermediate strengths possible for input with intermediate frequency levels. According to another embodiment, pixel classification determines the filtering that is applied to the dither. Pixels classified as low-frequency select an identity filter and thus, clustering dither is applied, providing stability. Pixels classified as high-frequency select a smoothing filter. The smoothing filter can attenuate the pattern of the dither. At the extreme, such smoothing would blur the dither into a flat, constant signal, and the resulting attenuated or smoothed dither produces less clustering.
According to yet another embodiment of the present invention, there is disclosed a method of selective dithering. The method includes receiving a plurality of input pixels of an original input image and identifying the pixel classification of the plurality of input pixels, where the pixel classification characterizes each of the plurality of pixels based, at least in part, on their respective gray values. The method also includes adding a dither input and at least some of the plurality of input pixels to generate a dithered image, where the strength of the dither input applied to generate the dithered image is based on the pixel classification of at least some of the plurality of input pixels.
According to one aspect of the present invention, adding a dither input and at least some of the plurality of input pixels includes adding a dither input and at least some of the plurality of input pixels using at least one interpolation table. Adding a dither input and at least some of the plurality of input pixels may alternatively include adding a dither input and at least some of the plurality of input pixels using at least one adder. According to another aspect of the present invention, the method further includes attenuating the dither input based on the pixel classification prior to adding the dither input and the at least some of the plurality of input pixels. According to yet another aspect of the present invention, the attenuation comprises a convolution filter whose coefficients are determined by the pixel classification of the plurality of input pixels.
According to another aspect of the present invention, the method includes performing error diffusion on the dithered image. Additionally, the pixel classification may be operable to characterize at least one of the plurality of pixels as an edge pixel or as a smooth pixel. Furthermore, the strength of the dither input applied to generating the dithered image may be lower for the edge pixel than for the smooth pixel.
According to another embodiment of the present invention, there is disclosed a method of applying dithering. The method includes determining the pixel classification of a plurality of input pixels in an original image and selecting, based upon the pixel classification, whether or not to add a dither input to at least some of the plurality of input pixels of the original image, where the dither input is not added to the at least some of the plurality of input pixels if the determined pixel classification is a first pixel classification, and the dither input is added to the at least some of the plurality of the input pixels if the determined pixel classification is a second pixel classification.
According to an aspect of the invention, the method may also include selectively attenuating the dither input based on the determined pixel classification. Additionally, selectively attenuating the dither input may be implemented by at least one filter. According to another aspect of the present invention, the method may also be performed by an application specific integrated circuit.
According to yet another embodiment of the invention, there is disclosed a system for applying dither. The system includes a dither input, a plurality of input pixels corresponding to an original input image, and a pixel classification module, where the pixel classification module is operable to receive the plurality of input pixels and further operable to characterize each of the plurality of pixels based, at least in part, on the respective gray values of neighboring pixels. The system may also include an adder, in communication with the dither input, where the adder is operable to add the dither input to at least some of the plurality of input pixels to generate a dithered image, where the strength of the dither input in generating the dithered image is based on the pixel classification of at least some of the plurality of input pixels.
According to one aspect of the invention, the adder includes one or more interpolation tables, which may be selected based on the pixel classification of at least some of the plurality of input pixels. According to another aspect of the present invention, the system may also include at least one smoothing filter operable to receive and attenuate the dither input based on the pixel classification of at least some of the plurality of input pixels prior to the adding of the dither input and the at least some of the plurality of input pixels. The system may also include at least one image filter operable to receive and filter the input pixels prior to the adding of the dither input and the at least some of the plurality of input pixels.
Having thus described the invention in general terms, reference will now be made to the accompanying drawings, which are not necessarily drawn to scale, and wherein:
The present invention will now be described more fully hereinafter with reference to the accompanying figures, in which like numerals indicate like elements throughout the several drawings. Some, but not all embodiments of the present invention are described. Indeed, the present invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use herein of “including,” “comprising” or “having” and variations thereof is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. The terms “mounted,” “connected” and “coupled” are used broadly and encompass both direct and indirect mounting, connecting and coupling. Further, “connected” and “coupled” are not restricted to physical or mechanical connections or couplings, and can include electrical connections or couplings, whether direct or indirect. The order of limitations specified in any method claims does not imply that the acts set forth therein must be performed in that order unless an order is explicitly set forth in the specification.
In addition, it should be understood that embodiments of the present invention include both hardware and electronic components or modules that, for purposes of discussion, may be illustrated and described as if the majority of the components were implemented solely in hardware. However, one of ordinary skill in the art, and based on a reading of this detailed description, would recognize that, in at least one embodiment, the electronic based aspects of the present invention may be implemented in software. As such, it should be noted that a plurality of hardware and software based devices, as well as a plurality of different structural components may be utilized to implement the present invention. Furthermore, and as described in subsequent paragraphs, the specific mechanical configurations illustrated in the drawings are intended to exemplify embodiments of the present invention, and other alternative mechanical configurations are possible.
Referring now to
According to the illustrative embodiment of the present invention shown in
As shown in
As shown in
According to an aspect of the present invention, the classification of the input pixels 105 may be accomplished by systems and methods disclosed in U.S. patent application Ser. No. 10/744,884, filed on Dec. 23, 2003, and titled “JPEG Encoding For Document Images Using Pixel Classification”, which is assigned to Lexmark International, Inc., the contents of which are incorporated by reference as if set forth fully herein. According to another aspect of the invention, the classification of pixels may be accomplished by systems and methods disclosed in U.S. patent application Ser. No. 10/224,660, filed on Aug. 20, 2002, titled “Systems and Methods For Content-Based Document Image Enhancement”, which is assigned to Lexmark International, Inc., the contents of which are incorporated by reference as if set forth fully herein. Collectively, the Ser. Nos. 10/744,884 and 10/224,660 patent applications will be referred to hereafter as the “Pixel Classification Applications.”
The pixel classification module 115 may incorporate any combination of hardware and/or software used to implement pixel classification as described in the Pixel Classification Applications. In particular, the pixel classification module 115 may classify each input pixel based on an analysis of neighboring pixels. According to one aspect, a feature vector can be developed to establish pixel classification, where the feature vector condenses the description of relevant properties of each pixel in the image into a small, Euclidean feature space. The number and type of elements included in the feature vector can be selected based on the complexity of the image. For instance, the elements of the feature vector may include, e.g., gray values, gray values obtained through different filters, texture measures, Markov random field features, fractal dimension measures, and gradient magnitudes and directions. In other words, the elements for a given pixel typically provide a measure of change occurring in the image at that pixel based on image data analyzed from neighboring pixels. The feature vector, and other values based thereon, may be used to classify each of the input pixels 105 using neighboring pixels, as described in detail in the Pixel Classification Applications. According to another aspect of the present invention, pixel classification may be based on each of, or one or more of, the color components for color pixels. Other pixel components in addition to gray scale and color components may also be used to classify pixels. For instance, derivatives of color components, such as luminance, may also be used for pixel classification.
Alternatively, as will be recognized, region based image segmentation can provide pixel classification information. Pixel classification may also be available a priori, such as generated from region based segmentation, or from information provided or based upon a display list representation of the image contents.
Referring once again to
As is also shown in
The mixer 125 represents multiple different interpolation tables that may be used to mix the input dither 110 with the filtered image received from the image filters 120. According to an aspect of the present invention, the mixer 125 represents two dimensional interpolation tables existing within an ASIC that implements the system 100. The contents of the two dimensional interpolation tables represent the sum of the dither input 110 and filtered image pixel inputs.
Depending on the pixel classification, and in particular, the pixel classification signal output of the pixel classification module 115, different interpolation tables may be used to effect dither of different strengths. As described above and as shown in
Although illustrated as interpolation tables, it will be appreciated by those of ordinary skill in the art that the mixer 125 may alternatively comprise adders to combine the dither input 110 with the filtered image output of the image filters 120. Multiple adders are necessary to alter the strength of mixing among the dither and input pixels. Nevertheless, because interpolation tables can allow amplitudes resulting from dither and pixel values to be constrained to the dynamic range of pixels while maintaining effectively zero mean dither, an equivalent arithmetic logic to the interpolation tables may be more complicated than a simple arithmetic adder, as will be appreciated by those of ordinary skill in the art.
As shown in FIG, 1, according to one embodiment, the output of the mixer 125 may be used as the input to error diffusion 130. The system 100 of
In the system 200 of
According to one aspect of the present invention, where the pixel classification module 215 determines pixel classification based, for example, on a 5 x 5 block of pixels with the center pixel of the block receiving classification, the dither filter modules 235 may be, for example, a 5 x 5 convolution filter with equal coefficients, where the center coefficient is applied to the pixel being filtered and the surrounding coefficients are correspondingly applied to surrounding pixels. Coefficients can be normalized so that they sum to unity; for example, each coefficient for the maximally smoothing 5×5 convolution filter is 1/25.
As described above, a feature vector can be developed to establish pixel classification. It will be appreciated that as the feature vector for pixel classification changes, differences in the feature vector resulting from edges or text input pixels can result in changes in the dither filter modules 235. For instance, where the dither filter modules 235 is a convolution filter, differences in the feature vector may cause incremental changes in 5×5 convolution filter coefficients, where the center coefficient will increase to 1 for a smooth image and the surrounding coefficient will decrease to 0, in which case the dither filter modules 235 applies no attenuation to the dither input 210.
It will be appreciated that whereas in
In the illustrative embodiment of
The general purpose processor 510 receives pixel data 506 from the main memory 504. The error diffusion processor 520 receives error spread coefficients 525 from the error spread coefficient system 530 and modified threshold information or error 543. The error diffusion processor 520 uses this information, along with previous line running error data 523 from an error buffer 550 to transform the pixel data 506 into error diffused pixel data 508 which is stored in the main memory 504. Control signals 521 are sent from the error diffusion processor 520 to the error spread coefficient system 530 for requesting coefficients and performing other functions. Alternative error spread coefficients may be provided based on pixel classification.
The error spread coefficient system 530 receives input 512 from the microprocessor 510. It may also receive pixel data 506 in the case of data-driven determinations of the error spread coefficients 525. The error spread coefficient system 530 provides the error diffusion processor 520 with the error spread coefficients 525 to be used in allocating the error from a transformed pixel. As discussed further below, the error spread coefficient system 530 may be implemented in a number of different ways.
The threshold generation logic 540 is under the control of the microprocessor 510 and may take on any number of forms. As seen in
The ASIC 502 of
Specifically, the first coefficient b1564 and the dither value D[x,y] 562 are input to a first multiplier 576, and the first product from this first multiplier 576 is forwarded to first adder 572. At first adder 572, this first product is added to the input pixel value Io[x,y] 506 to form the dithered pixel data In[x,y] 507. Preferably, first coefficient b1 is less than 1.0 so that the output of multiplier 576 (i.e., the passband dither) has a smaller dynamic range than original input pixel values Io[x,y] 506. The second coefficient b2566 and the dither value D[x,y] 562 are input to a second multiplier 578, and this second product is forwarded to second adder 574. At second adder 574, this second product is added to the tone-dependent threshold value Tn[Io[x,y]] 542 to produce modified threshold value Tm[x,y] (i.e., error) 543, which is then input to the error diffusion processor 520.
It is noted that while ASIC 502 shows that the original input pixel data Io[x,y] 506 is used to index the threshold values in the threshold generation logic 540, one may instead connect the dithered pixel data In[x,y] 507 to the threshold generation logic 540 to thereby use information from the dithered input pixel values to index the threshold values. It is also noted that in the ASIC implementation of
Furthermore, the microprocessor, or other hardware, can be configured to produce or otherwise provide the various passband dither signals discussed above; look for edges in a region proximate to a pixel of interest and, if appropriate, suspending addition of the first type of passband dither; and determine an average pixel value in a region proximate to a pixel of interest and, if appropriate, take any necessary action, all as discussed above. In the implementation of ASIC 502, the dithered input pixel values In[x,y] 507 are computed with hardware adders and multipliers. For example, the microprocessor 510 may perform the entire computation in software and provide dithered input pixel values In[x,y] 507 to the error diffusion processor 520. Alternatively, in one embodiment, the dither values D[x,y] 562, already scaled by b1, may be stored in a lookup table, indexed by the pixel [x,y] position, and accessible to the microprocessor 510 which then adds the retrieved information to the input pixel value 506 and again provides the result to the error diffusion circuitry. Such a lookup table may form a portion of the microprocessor's local memory. The same principles hold for producing the modified threshold values Tm[x,y] 543. The ASIC 502 may be configured in some manner to add the proper first type of dither to the input pixel values, and add the proper second type of dither to the threshold values. In the embodiment shown, the first type of passband dither and the second type of passband dither have the same spectral characteristics and differ by a scalar multiple.
Finally, as is known to those skilled in the art, the error diffusion processor 520 typically processes image pixel data in line-order—each pixel belonging to one line of an image is error diffused, before pixels of the next line are processed. The error buffer 550 receives next line error data 522 from the error diffusion processor 520 for use in distributing the error from an immediately preceding row of image data and provides these errors back 523 to the error diffusion processor 520 for use in adjusting a current pixel value of a pixel in a current row of image data. It is understood that in some prior art embodiments, the error buffer 550 may be instantiated in main memory 504 when memory bandwidth suffices to support the pixel rate. In other prior art embodiments, circuitry for implementing the error buffer 550 may simply be integral to the error diffusion processor 520.
While in the prior art embodiment of
Many modifications and other embodiments of the present invention set forth herein will come to mind to one skilled in the art to which this invention pertains having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the present invention is not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
This application claims priority to U.S. provisional patent application No. 60/709,914 titled, “Systems and Methods for Selective Dithering Using Pixel Classification,” which was filed on Aug. 18, 2005.
| Number | Date | Country | |
|---|---|---|---|
| 60709914 | Aug 2005 | US |
