This PDF file contains the front matter associated with the Proceedings of SPIE Volume 6493, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

There is a growing family of algorithms that treat/modify/enhance color information in its visual context, also known as Spatial Color methods (e.g. Retinex or ACE). They produce results that, due to a changing spatial configuration, can have a non-unique relationship with the physical input. In authors' opinion judging their performance is a challenging task and is still an open problem. Two main variables affect the final result of these algorithms: their parameters and the visual characteristics of the image they process. The term visual characteristics refers not only to the image's digital pixel values, (e.g. calibration of pixel value, the measured dynamic range of the scene, the measured dynamic range of the digital image), but also to the spatial distribution of these digital pixel values in the image. This paper does not deal with tuning parameters, rather it discusses the visual configurations in which a Spatial Color methods show interesting, or critical behavior. A survey of the more significant Spatial Color configurations will be presented and discussed. These configurations include phenomena, such as color constancy and contrast. The discussion will present strengths and weaknesses of different algorithms, hopefully allowing a deeper understanding of their behavior and stimulating discussions about finding a common judging ground.

One of latest developments for pre-press applications is the concept of soft proofing, which aims to provide an accurate preview on a monitor of how the final document will appear once it is printed. At the core of this concept is the problem of identifying, for any printed color, the most similar color the monitor can display. This problem is made difficult by such factors as varying viewing conditions, color gamut limitations, or the less studied time spacing. Color matching experiments are usually done by examining samples viewed simultaneously. However, in soft proofing applications, the proof and the print are not always viewed together. This paper attempts to shed more light on the difference between simultaneous and time-spaced color matching, in order to contribute to improving the accuracy of soft proofs. A color matching experiment setup has been established in which observers were asked to match a color patch displayed on a LCD monitor, by adjusting its RGB values, to another color patch printed out on paper. In the first part of the experiment the two colors were viewed simultaneously. In the second part, the observers were asked to produce the match according to a previously memorized color. According to the obtained results, the color appearance attributes lightness and chroma were the most difficult components for the observers to remember, generating huge differences with the simultaneous match, whereas hue was the component which varied the least. This indicates that for soft proofing, getting the hues right is of primordial importance.

Mixed Raster Content (MRC) is a standard for efficient document compression which can dramatically improve the compression/quality tradeoff as compared to traditional lossy image compression algorithms. The key to MRC's performance is the separation of the document into foreground and background layers, represented as a binary mask. Typically, the foreground layer contains text colors, the background layer contains images and graphics, and the binary mask layer represents fine detail of text fonts. The resulting quality and compression ratio of a MRC document encoder is highly dependent on the segmentation algorithm used to compute the binary mask. In this paper, we propose a novel segmentation method based on the MRC standards (ITU-T T.44). The algorithm consists of two components: Cost Optimized Segmentation (COS) and Connected Component Classification (CCC). The COS algorithm is a blockwise segmentation algorithm formulated in a global cost optimization framework, while CCC is based on feature vector classification of connected components. In the experimental results, we show that the new algorithm achieves the same accuracy of text detection but with lower false detection of non-text features, as compared to state-of-the-art commercial MRC products. This results in high quality MRC encoded documents with fewer non-text artifacts, and lower bit rate.

The JBIG2 binary image encoder dramatically improves compression ratios over previous encoders. The effectiveness of JBIG2 is largely due to its use of pattern matching techniques and symbol dictionaries for the representation of text. While dictionary design is critical to achieving high compression ratios, little research has been done in the optimization of dictionaries across stripes and pages. In this paper we propose a novel dynamic dictionary design that substantially improves JBIG2 compression ratios, particularly for multi-page documents. This dynamic dictionary updating scheme uses caching algorithms to more effciently manage the symbol dictionary memory. Results show that the new dynamic symbol caching technique outperforms the best previous dictionary construction schemes by between 13% and 46% for lossy compression when encoding multi-page documents. In addition, we propose a fast and low-complexity pattern matching algorithm that is robust to substitution errors and achieves high compression ratios.

We present a segmentation-based post-processing method to remove compression artifacts from JPEG compressed document images. JPEG compressed images typically exhibit ringing and blocking artifacts, which can be objectionable to the viewer above certain compression levels. The ringing is more dominant around textual regions while the blocking is more visible in natural images. Despite extensive research, reducing these artifacts in an effective manner still remains challenging. Document images are often segmented for various reasons. As a result, the segmentation information in many instances is available without requiring additional computation. We have developed a low computational cost method to reduce ringing and blocking artifacts for segmented document images. The method assumes the textual parts and pictorial regions in the document have been separated from each other by an automatic segmentation technique. It performs simple image processing techniques to clean out ringing and blocking artifacts from these regions.

In the archiving and museum communities, the long-term preservation of artworks has traditionally been guaranteed by making duplicates of the original. For photographic reproductions, digital imaging devices have now become standard, providing better quality control and lower costs than film photography. However, due to the very short life cycle of digital data, losses are unavoidable without repetitive data migrations to new file formats and storage media. We present a solution for the long-term archiving of digital images on color microfilm (Ilfochrome® Micrographic). This extremely stable and high-resolution medium, combined with the use of a novel laser film recorder is particularly well suited for this task. Due to intrinsic limitations of the film, colorimetric reproductions of the originals are not always achievable. The microfilm must be first considered as an information carrier and not primarily as an imaging medium. Color transformations taking into account the film characteristics and possible degradations of the medium due to aging are investigated. An approach making use of readily available color management tools is presented which assures the recovery of the original colors after re-digitization. An extension of this project considering the direct recording of digital information as color bit-code on the film is also introduced.

This paper describes an investigation into the effect of a wide range of surround conditions on the colour appearance of test colours on a 42" plasma display panel. Experiments were conducted using surrounds including dark, indoor and outdoor conditions. Additionally the stimulus size was changed by controlling the viewing distance. The viewing conditions studied were two bright, two average, two dim and two dark surrounds. Each of the test colours was assessed by 10 observers using a magnitude estimation method. These surrounds were divided into two categories. In the first category, the surround had no effect on the displayed colours, but observers could still sense the different brightness levels of the surround. In the second category, surround introduced flare to the displayed colours together. For the first category, little visual lightness difference was shown between bright and dark, and dim and dark surround, unlike the expectation that the perceived lightness contrast may increase as the surround becomes brighter. The lightness dependency of colourfulness, however, was found to change. For the second category, the visual colour appearance of the surround conditions was plotted against measured data, CIELAB L*, C* values, to try to understand the surround effect. As the surround became brighter, the perceived dynamic range of visual lightness decreased, and the perceived colourfulness increased, more obviously in high chroma colours. In the investigation of the change of stimulus size under different surround conditions, visual colour appearance was not affected by the stimulus sizes of 2^o and 0.6^o in the dark surround. However, the difference was found in the very dark colours with a dim surround. Finally, all of visual colour appearance data were used to test the performance of the colour appearance model CIECAM02. Minor modification was accomplished to improve the colourfulness predictor, especially for the black background.

Many art objects have a size much larger than their softcopy reproductions. In order to develop a multiscale model that accounts for the effect of image size on image appearance, a digital projector and LCD display were colorimetrically characterized and used in a contrast matching experiment. At three different sizes and three levels of contrast and luminance, a total of 63 images of noise patterns were rendered for both displays using three cosine log filters. Fourteen observers adjusted mean luminance level and contrast of images on the projector screen to match the images displayed on the LCD. The contrasts of the low frequency images on the screen were boosted while their mean luminance values were decreased relative to the smaller LCD images. Conversely, the contrast of projected high frequency images were reduced for the same images on LCD with a smaller size. The effect was more pronounced in the matching of projected image to the smaller images on the LCD display. Compared to the mean luminance level of the LCD images, a reduction of the mean luminance level of the adjusted images was observed for low frequency noise patterns. This decrease was more pronounced for smaller images with lower contrast and high mean luminance level.

This paper proposes a color correction method based on modeling the hue shift phenomenon of human visual system (HVS). Observers tend to perceive same color stimuli, but of different intensity, as different hues, what is referred to as the hue shift effect. Although the effect can be explained with the Bezold-Brücke (B-B) effect, it is not enough to apply the B-B model on high luminance displays because most displays have a broad-band spectrum distribution and results vary according to type of display. In the proposed method, the quantities of hue shift between a high luminance display and a normal luminance display were first modeled by a color matching experiment with color samples along the hue angle of the LCH color space. Based on the results, the hue shift was then modeled piecewise and was finally applied to the inverse characterization of display to compensate the original input image. From evaluating the proposed method using the psychophysical experiment with some test images, we confirmed that the proposed modeling method is effective for color correction on high luminance displays.

Large-scale, direct view TV screens, in particular those based on liquid crystal technology, are beginning to use structures with more than three subpixels to further reduce the visibility of defective subpixels, increase spatial resolution, and perhaps even implement a multi-primary display with up to six different primaries. This newly available "subpixel" resolution enables us to improve color edge contrast thereby allowing for shorter viewing distances through a reduction of perceived blur. However, not only noise, but also amplitude quantization can lead to undesirable, additional visual artifacts along contours. Using our recently introduced method of contour phase synthesis in combination with non-linear, color channel processing, we propose a simple method that maximizes color edge contrast while maintaining or suppressing visual artifacts as well as noise below the threshold of visual perception. To demonstrate the advantages of our method, we compare it with some classical contrast enhancement techniques such as cubic spline interpolation and color transient improvement.

The present study provides functions to correlate gamut size across different color spaces including 2D planes - (x,y) and (u',v') and 3D spaces - CIELAB, CIECAM02 JCh and QMh. All gamut size must be converted to NTSC gamut ratio before using the functions. As viewing conditions influence 3D gamut ratio significantly, predict 3D gamut ratio in high precision is not easy. However, the mean values, median values, standard deviations, and confidence intervals can be predicted accurately. In terms of viewing parameters, we can model them individually under IEC reference condition successfully, the resulted functions would be a good reference to derive more versatile functions to predict gamut ratio under complex viewing conditions.

Due to subtle misalignment of optical components in the fabrication process, images projected by an optical light modulator have severe line artifact along the direction of the optical scan. In this paper, we propose a novel methodology to calibrate the modulator and generate the compensate image for the misaligned optical modulator in order to eliminate the line artifact. A camera system is employed to construct Luminance Transfer Function (LTF) that characterizes the optical modulator array. Spatial uniformity is obtained by redefining the dynamic range and compensating the characteristic curvature of the LTF for each optical modulator array element. Simulation results show significant reduction in the visibility of line artifact.

The world's best motion fidelity LCD, which is based on scanning backlight technique, was developed. The MPRT (Motion Picture Response Time) of 4.0 ms for an LCD is achieved while a CRT has about 3.7 ms and a PDP still has about 8.1 ms. It means that the last weakness bothering an LCD so far compared with a PDP and a CRT was cleared. Moreover, 4.0 ms MPRT is a number at which people cannot perceive motion blur for moving images under most conditions and therefore the MBR LCD is superior to a PDP by above 50% and equal to a CRT in motion blur.

High dynamic compression of natural scenes and nonlinearity compensation of output devices are demanded as the output devices have limited dynamic range and color gamut, and devices such as Standard-RGB compliant displays and printers have different nonlinear responses to linear inputs. This paper describes a general framework of nonlinear curve optimization in digital imaging and its application to the determination of gamma look up table through curve parameterization. Three factors are considered, that is, the color appearance fidelity, the feature preservation, and the suppression of noise propagation.

The paper presents two models to reduce the visibility of rainbow effect for field sequential color (FSC) LCDs. It was done by changing the LED backlight state, introducing crosstalk among the R/G/B LEDs and finally modified the RGB signals to minimize its image color differences to its original state. The results of our simulations suggested that the proposed methods would reduce the visibility of rainbow effect to some extent. However, its performance was quite image-dependent. How to reduce its computational cost and calibrate the LED backlight accurately is a great challenge to implement the models on real FSC LCDs.

We can all use, to a larger or lesser extent, the technological marvels that surround us and that we often depend on; without, however, really understanding how they function. But designing, buying, using and maintaining those devices does require a certain level of knowledge on the basics of that functioning, if one wants to avoid problems. Displays are a worth-while example of the marvels meant because of their ubiquity in modern life. A standard on displays, especially a new one, could provide the knowledge meant - and anyway deserves to be known by the people frequenting the EI Conferences. Therefore, a progress report on the production of the new ISO display standard is presented here for the third time. The standard contains: an introduction; all definitions used; the visual requirements for displays; three different methods to measure display properties; and analysis and compliance methods to verify whether a particular display complies with the requirements set. The standard is meant to provide knowledge for designers of displays, their users, and those that procure displays for these users. It is set up in such a way that emerging display technologies as are used in, for instance, SED and OLED, can be incorporated easily.

We propose a novel color mapping method that generates smooth color transition and can accommodate the color preference. The method consists of two stages; rough calibration and black generation. The rough calibration process generates a three dimensional (3-D) Look-Up-Table converting input RGB data to printable CMY values. When the 3-D LUT is created, a new intent for color mapping, target color is internally used. The target color is predefined from a reference color book based on the color preferences of testers involved in the target definition phase. All of the input data of the 3-D LUT are mapped to the printable values in a printer based on the target color, and then simply converted to CMYK values. We evaluated the proposed algorithm comparing with a commercial printer profiler and found that the proposed algorithm makes better printing quality.

The same image on the display and color printer isn't the same. Firstly, this is due to the bit depth difference for representing the color of a pixel. The display uses the color data of the eight or more bits, but the color printer uses just 1bit for representing color of a pixel. So, the display can reproduce smoother image than the color printer. Secondly, the display gamut is larger than the printer gamut, so the display color is brighter and more saturate than the printer color. For minimizing the problems due to these differences, many halftoning and gamut mapping techniques have been developed. For the gamut mapping, color management standard organization, ICC, recommended 2 gamut mapping methods, HPMINDE and SGCK. But the recommended methods by ICC have some weak points; contouring (HPMINDE), paled pure color reproduction (SGCK) and too reddish hair color reproduction (HPMINDE, SGCK). This paper introduces a gamut mapping method that can reproduce smooth gradation, pure colors with high saturation and natural hair color. The proposed method is developed for optimal reproduction of graphic image, and it also gives good results for pictorial image.

The words in a document are often supported, illustrated, and enriched by visuals. When color is used, some of it is used to define the document's identity and is therefore strictly controlled in the design process. The result of this design process is a "color specification sheet," which must be created for every background color. While in traditional publishing there are only a few backgrounds, in variable data publishing a larger number of backgrounds can be used. We present an algorithm that nudges the colors in a visual to be distinct from a background while preserving the visual's general color character.

Redeyes are caused by the camera flash light reflecting off the retina. Peteyes refer to similar artifacts in the eyes of other mammals caused by camera flash. In this paper we present a peteye removal algorithm for detecting and correcting peteye artifacts in digital images. Peteye removal for animals is significantly more difficult than redeye removal for humans, because peteyes can be any of a variety of colors, and human face detection cannot be used to localize the animal eyes. In many animals, including dogs and cats, the retina has a special reflective layer that can cause a variety of peteye colors, depending on the animal's breed, age, or fur color, etc. This makes the peteye correction more challenging. We have developed a semi-automatic algorithm for peteye removal that can detect peteyes based on the cursor position provided by the user and correct them by neutralizing the colors with glare reduction and glint retention.

This paper will outline a simplified model for the development of toner dots on a reflective support. Using this model and the interaction of light between the reflective support and the dot's microstructure, the physical, optical and total dot-gain will be calculated, along with the resulting tone scales, for a variety of digital halftone patterns. The resulting tone reproduction curves and dot-gain will be compared with the classical literature on dot-gain and tone reproduction curves, more modern approaches and experimental data from the literature. A comparison to a well-defined experimental system will be shown.

This paper will outline a simplified model for the development of toner dots on a reflective support in a color electro-photographic system. A model developed for a monochrome system will be adapted to a color imaging system where four pigments, each capable of scatting light, is used to form a digital halftone image. The combination of physical and optical dot gains, interlayer scattering, on-dot and off-dot digital halftones will be explored and the results demonstrated in terms color shifts due to layer order and dot gain due to halftone geometry.

The purpose of printer color calibration is to ensure accurate color reproduction and to maintain the color consistency. A closed loop color calibration system has the ability to perform the whole calibration work automatically without any user intervention. A color calibration process consists of several steps: print a pre-designed target; measure the printed color patches using an color measuring instrument; run a set of algorithms to calibrate the color variations. In this paper, we present a closed loop color calibration system. And we will show in particular how to use a low cost optical sensor to get accurate color measurement. Traditional, low cost optical sensors are only used to measure the voltage data or density data. The novelty of our approach is that we can also use a low cost optical sensor to measure colorimetric data. Using the colorimetric measurement, we can perform more complicate color calibration works for color printing systems.

Papers with a slightly blue shade are, at least among a majority of observers being perceived as whiter than papers having a more neutral color1. Therefore, practically all commercially available printing papers contain bluish dyes and fluorescent whitening agents (FWA) to give the paper a whiter appearance. Furthermore, in the paper industry, the most frequently used measure for paper whiteness is the CIE-whiteness. The CIE Whiteness formula, does in turn, also favor slightly bluish papers. Excessive examples of high CIE-whiteness values can be observed in the office-paper segment where a high CIE-whiteness value is an important sales argument. As an effect of the FWA, spectrophotometer measurements of optical properties such as paper whiteness are sensitive to the ultraviolet (UV) content of the light source used in the instrument. To address this, the standard spectrophotometers used in the paper industry are equipped with an adjustable filter for calibrating the UV-content of the illumination. In the paper industry, spectrophotometers with d/0 measurement geometry and a light source of type C are used. The graphical arts industry on the other hand, typically measures with spectrophotometers having 45/0 geometry and a light source of type A. Moreover, these instruments have only limited possibilities to adjust the UV-content by the use of different weighting filters. The standard for color measurements in the paper industry governs that measurements should be carried out using D65 standard illumination and the 10^o standard observer. The corresponding standard for the graphic arts industry specify D50 standard illumination and the 2^o standard observer. In both cases, the standard illuminants are simulated from the original light source by spectral weighting functions. However, the activation of FWA, which will impact the measured spectral reflectance, depends on the actual UV-content of the illumination used. Therefore, comparisons between measurements on substrates containing FWA from two instruments having light sources with different UV-content are complicated. In this study, the effect of FWA content in paper on color reproduction has been quantified for an officetype paper. Furthermore, examples are given on how color measurement instruments give different readings when FWA is present. For the purpose of this study and in order to ensure that only the effect of FWA was observed, a set of papers with varying additions of FWA otherwise identical, were produced on a small-scale experimental paper machine. The pilot papers were printed in three different printers. Two spectrophotometers representative to the instruments used in the Graphical Art Industry and the Paper Industry respectively where used to measure the printed papers. The results demonstrate how the use of spectral weighting functions for simulating standard illuminants works properly on nonfluorescent material. However, when FWA is present, disparities in UV content between the light source and the simulated illuminant will result in color differences. Finally, in many printing processes, some of the used inks are UVblocking, this further complicates the effect of FWA in printed material. An example is shown on how different color differences are obtained for different process ink combinations when the amount of FWA added to the paper is varied.

A colour management strategy based on a reference printer and reference medium is described. Additional printers and media are then defined in terms of this reference printer. This makes it possible to estimate new transforms for a given a printer and media with no new measurements, even when there is no characterization data available for this combination of printer and media. A simple method of estimating the new transform was implemented and evaluated, and it was found to give good results when the target printing condition was similar in gamut and colorant primaries to the reference printer. This approach can be extended by applying more complex empirical models.

In order to print accurate colors on different substrates, color profiles must be created for each specific ink-media combination. We tackled the problem of creating such color profiles from only few color samples, in order to reduce the needed time of operation. Our strategy is to use a spectral reflectance prediction model in order to estimate a large sampling target (e.g. IT8.7/3) from only a small subset of color patches. In particular, we focused on the so-called Yule-Nielsen modified Spectral Neugebauer model, proposing new area coverage estimation, and a prediction of Neugebauer primaries, which can not be directly measured due to ink limiting. We reviewed the basis of such model, interpret it under the perspective of generalized averaging, and derived expressions to decouple optical and mechanical dot gain effects. The proposed area coverage estimations are based on assumptions of the printing process, and characterized through few extra color samples. We tested the models with thermal ink-jet printers on a variety of media, with dye-based and pigment-based inks. The IT8.7/3 target was predicted from 44 samples, with color average accuracy below 4 dE and maximum error below 8 dE, for dye-based inks, which performed better than pigment-based inks.

Flatbed scanners and digital cameras have become established and widely used color imaging devices. If colorimetrically calibrated, these trichromatic devices can provide fast color measurement tools in applications such as printer calibration, process control, objective print quality measurements and color management. However, in calibrations intended to be used for color measurements on printed matter, the media dependency must be considered. Very good results can be achieved when the calibration is carried out on a single media and then applied for measurements on the same media, or at least a media of a very similar type. Significantly poorer results can be observed when the calibration is carried out for one printer-substrate combination and then applied for measurements on targets produced with another printer-substrate combination. Even if the problem is restricted to the color calibration of a scanner or camera for different paper media printed on a single printer, it is still tedious work to make a separate calibration for each new paper grade to be used in the printer. Therefore, it would be of interest to find a method where it is sufficient to characterize for only one or a few papers within a grade segment and then be able to apply a correction based on measurable optical paper properties. However, before being able to make any corrections, the influence of measurable paper properties on color characterizations must be studied and modeled. Fluorescence has been mentioned1-3 as a potential source of error in color calibrations for measurements on printed matter. In order to improve paper whiteness, producers of printing paper add bluish dye and fluorescent whitening agents (FWA) to the paper4. In this study, the influence of FWA in printing paper on the color calibration of a digital camera for color measurements on printed targets is discussed. To study the effect of FWA in the paper, a set of papers with varying additions of FWA but otherwise identical, were produced on a smallscale experimental paper machine. Firstly, the impact on the color calibration when the amount of FWA in the paper varies was studied. Secondly, the situation where the printed substrate has FWA-content, and illuminations having different contents of ultraviolet (UV) light were used in the camera and reference spectrophotometer measurements respectively. The results show that for some combinations of illuminations used in the calibration, very large errors are induced by the variation of FWA in the printed substrate.

In order to obtain a good quality image in preparation for inkjet printing, the process of adjusting images can be a time consuming and a costly procedure. In this paper, we consider the use of an unsupervised colour enhancement method as part of the automatic pre-processors for printing. Other unsupervised colour enhancement methods are utilised and compared: Retinex, RSR, ACE, Histogram Equalisation, Auto Levels. Test images are subjected to all of the enhancement methods, which are then printed. Users are asked compare each of the sampled images. In all cases, the results are dependent on the image. Thus, we have selected a range of test images: photographs of scenes, reproduction of prints, paintings and drawings. Some of the tested methods are parameter dependent. We do not intend to consider fine tuning for each of the techniques, rather to consider an average parameter set for each one and then test if this approach can aid the decision process of fine tuning. Three user groups are employed: the general user, commercial photographer expert and fine artist. Groups are asked to make a blind evaluation of a range of images (the original and the colour enhanced by the different methods); these are randomly placed. All images are printed on the same printer using the same settings. Users are asked to identify their preferred print in relation to lightness, tonal range, colour range, quality of detail and overall subjective preference.

A huge number of consumer digital images are considered unsatisfactory for general use due to their poor image quality. Of this number a significant fraction actually possess an inherent quality, as acquired, and with relatively straightforward manipulation could be enhanced and thus become entirely satisfactory according to consumer preference. In fact around fifty percent of all digital images, including those acquired with high signal-to-noise ratio using sophisticated capture devices, are positioned at a significant distance from their optimum preferred image-quality state, as defined by the end user. An unconventional practical methodology has been developed that uses a novel image processing procedure, yet leads to a solution that can readily be applied in real-time by any unskilled consumer to obtain the preferred image-quality state for each individual digital image. The underlying scientific principles are now explored in further detail, and these are related to the practical image-quality outcomes. Examples are given of the transformations applied to specific image types in order to yield preferred image-quality, and these transformations are related directly to the selection process as operated by an unskilled user of the methodology.

The production planning in a printing organization is quite complex since there are many parameters to consider such as the work to be done, the available devices and the available resources. Printed products such as books, magazines, leaflets etc. all consist of sections that will be printed on a press sheet. After printing, the sheets will be folded and cut. As a last step, the different sections belonging to the same product will be collected and bound together (glued, stapled, stitched etc.). In the prepress environment, one traditionally uses special imposition templates to identify how the pages should be imposed on the press sheet. The main drawback of this approach is that one needs to remake imposition templates each time a parameter has been changed. A new approach to overcome this problem has been proposed by the CIP4 graphic standards organization and consists of specifying a so-called stripping specification of the sections on the sheet. In addition to the stripping information, one also can specify how the sections can be combined physically. This is done in the so-called assembly specification. Both stripping and assembly allow defining unambiguously how a product can be produced. In the first part of the paper, we will explain in detail how the stripping and assembly specification can be used to further automate the prepress, printing and finishing processes. In the second part, we will discuss how the production planning itself can be automated. This assumes an in-depth knowledge of the device characteristics (press and finishing equipment) and the available resources (plates, paper, ink).

An ever increasing amount of digital images are being captured. This increase is due to several reasons. People are afraid of not "capturing the moment" and pressing the shutter is not directly liked to costs as was the case with silver halide photography. This behaviour seems to be convenient but can result in a dilemma for the consumer. This paper presents tools designed to help the consumer overcome the time consuming image selection process while turning the chore of selecting the images for prints or placing them automatically into a photo book into a fun experience.

We introduce a novel algorithm for local contrast enhancement. The algorithm exploits a background image which is estimated with an edge-preserving filter. The background image controls a gain which enhances important details hidden in underexposed regions of the input image. Our designs for the gain, edge-preserving filter and chrominance recovery avoid artifacts and ensure the superior image quality of our results, as extensively validated by user evaluations. Unlike previous local contrast methods, ours is fully automatic in the sense that it can be directly applied to any input image with no parameter adjustment. This is because we exploit a trainable decision mechanism which classifies images as benefiting from enhancement or otherwise. Finally, a novel windowed TRC mechanism based on monotonic regression ensures that the algorithm takes only 0.3 s to process a 10 MPix image on a 3 GHz Pentium.

This paper proposes a colorization method that uses wavelet packet sub-bands to embed color components. The proposed method, firstly, involves a color-to-gray process, in which an input RGB image is converted into Y, Cb, and Cr images, and a wavelet packet transform applied to Y image to divide it into 16 sub-bands. The Cb and Cr images are then embedded into two sub-bands that include minimum information on the Y image. Once the inverse wavelet packet transform is carried out, a new gray image with texture is obtained, where the color information appears as texture patterns that are changed according to the Cb and Cr components. Secondly, a gray-to-color process is performed. The printed textured-gray image is scanned and divided into 16 sub-bands using a wavelet packet transform to extract the Cb and Cr components, and an inverse wavelet packet transform is used to reconstruct the Y image. At this time, the original information is lost in the color-to-gray process. Nonetheless, the details of the reconstructed Y image are almost the same as those in the original Y image because it uses sub-bands with minimum information to embed the Cb and Cr components. The RGB image is then reconstructed by combining the Y image with the Cb and Cr images. In addition, to recover color saturations more accurately, gray patches for compensating the characteristics of printers and scanners are used. As a result, the proposed method can improve both the boundary details and the color saturations in recovered color images.

The Yule-Nielsen modified Spectral Neugebauer reflection prediction model enhanced with an ink spreading model provides high accuracy when predicting reflectance spectra from ink surface coverages. In the present contribution, we try to inverse the model, i.e. to deduce the surface coverages of a printed color halftone patch from its measured reflectance spectrum. This process yields good results for cyan, magenta, and yellow inks, but unstable results when simultaneously fitting cyan, magenta, yellow, and black inks due to redundancy between these four inks: black can be obtained by printing either the black ink or similar amounts of the cyan, magenta, and yellow inks. To overcome this problem, we use the fact that the black pigmented ink absorbs light in the infrared domain, whereas cyan, magenta, and yellow inks do not. Therefore, with reflection spectra measurements spanning both the visible and infrared domain, it is possible to accurately deduce the black ink coverage. Since there is no redundancy anymore, the cyan, magenta, yellow, and pigmented black ink coverages can be recovered with high accuracy.

In imaging systems, color plays an essential role in conveying and recording visual information from the real world. To faithfully represent colors acquired from digital cameras, a spectral responsivity measurement system is proposed for those devices in this paper. For estimating spectral responsivities of digital color cameras, a filter-based optical system is designed with proper filter selections. Since the spectral filters primarily prescribe the optical characteristics of the system, the filter consideration is important to the optical design of the system with the presence of noise. A theoretical basis is presented to confirm that sophisticated filter selections can make this system as insensitive to noise as possible. Also, we propose a filter selection method based on the orthogonal-triangular (QR) decomposition with column pivoting (QRCP). To investigate the noise effects, we assess the estimation errors between the actual and estimated spectral responsivities, with the different signal-to-noise ratio (SNR) levels of an eight-bit/channel camera. To demonstrate the effectiveness of this approach, the experimental results from the filter-based optical system with the spectral filters selected from the QRCP-based method is much less sensitive to noise than those with other filters from different selections. It is found that the measurement accuracy is fairly satisfactory.

In this paper we propose a method to speed up Retinex-type algorithms, consisting of a computationally intensive nonlinear illumination estimation module followed by a relatively simple manipulation module. Speed up is obtained by way of computing the illumination on a sub-sampled image. The challenge is to interpolate a piece-wise smooth low resolution image. We present and analyze the trade of between two types of interpolation methods. On one hand, regular illumination interpolation, which preserves the Retinex-type output quality, however may result in artifacts. On the other hand a detail preserving interpolation which removes artifacts, however may compromise the output quality.

A method is developed for estimating an omnidirectional distribution of the scene illuminant spectral distribution, including spiky fluorescent spectra. First, we show a measuring apparatus, consisting of the mirrored ball system and the imaging system using a LCT filter (or color filters), a monochrome CCD camera, and a personal computer. Second, the measuring system is calibrated and images representing the omnidirectional light distribution are created. Third, we present an algorithm for recovering the illuminant spectral-power distribution from the image data. Finally, the feasibility of the proposed method is demonstrated in an experiment on a classroom scene with different illuminant sources such as fluorescent light, incandescent light, and daylight. The accuracy of the estimated scene illuminants is shown in the cases of the 6-channel multi-band camera, 31-channel spectral camera, and 61-channel spectral camera.

The color of prints is mainly determined by the light absorption of the inks deposited on top of paper. In order to predict the reflectance spectrum of prints, we use a spectral prediction model in which each ink is characterized by its spectral transmittance. In the present paper, we consider two classical reflectance prediction models: the Clapper-Yule model and the Williams-Clapper model. They rely on a same description of multiple reflection-transmission of light, but use a different description of the attenuation of light by the inks. In the Clapper-Yule model (non-orientational ink attenuation), the orientation of light traversing the ink is not taken into account. In the Williams-Clapper model, it is taken into account (orientational ink attenuation). In order to determine experimentally which of these two models is the more suitable for a given type of print, we propose a method using the reflectance and the transmittance of prints. We introduce a bimodal model, enabling spectral reflectance and transmittance predictions. Depending whether the direction of light into the ink is taken into account, we obtain a non-orientational bimodal model or an orientational bimodal model. Using these two models, we deduce the ink transmittance spectrum from various reflectance and transmittance measurements performed on a same print, and compare the different deduced spectra. The model which is the most adapted to the considered print is the one where the deduced spectra best match each other.

Color printer calibration is the process of deriving correction functions for device CMYK signals, so that the device can be maintained with a fixed known characteristic color response. Since the colorimetric response of the printer can be a strong function of the halftone, the calibration process must be repeated for every halftone supported by the printer. The effort involved in the calibration process thus increases linearly with the number of halftoning methods. In the past few years, it has become common for high-end digital color printers to be equipped with a large number of halftones thus making the calibration process onerous . We propose a halftone independent method for correcting color (CMY/CMYK) printer drift. Our corrections are derived by measuring a small number of halftone independent fundamental binary patterns based on the 2×2 binary printer model by Wang et. al. Hence, the required measurements do not increase as more halftoning methods are added. The key novelty in our work is in identifying an invariant halftone correction factor (HCF) that exploits the knowledge of the relationship between the true printer response and the 2×2 predicted response for a given halftoning scheme. We evaluate our scheme both quantitatively and qualitatively against the printer color correction transform derived with the printer in its "default state". Results indicate that the proposed method is very successful in calibrating a printer across a wide variety of halftones.

We aim to print spectral images using spectral vector error diffusion. Vector error diffusion produces good quality halftoned images but it is very slow to diffuse the error in the image during the halftoning process due to error accumulation. In spectral images each pixel is a re.ectance and the accumulation of error can modify completly the shape of the reflectance. This phenomena is increased when data are out of the gamut of the printer. To control the diffusion of error and to decrease the slowness of the spectral vector error diffusion we preprocess the spectral image by applying spectral gamut mapping and test the shape of the reflectances by keeping them in a range of feasible values. Our spectral gamut mapping is based on the inversion of the spectral Neugebauer printer model. After preprocessing the spectral image to be halftoned is the closest estimation the printer can made of it with the available colorants. We apply spectral vector error diffusion to spectral images and we evaluate the halftoning by simulation. We use a seven channels printer which we assume has stable inks and no dot gain (with a large set of inks we increase the variability or re.ectances the printer can produce). Our preprocessing and error control have shown promising results.

A new method for halftoning using high resolution pattern templates is described, that expands the low level rendering capabilities for engines that support this feature. This approach, denoted super resolution encoded halftoning (SREH) is an extension of the Holladay concept, and provides a compact way to specify high resolution dot growth patterns using a lower resolution Holladay brick. Fundamentally, this new halftoning method involves using the SRE patterns as building blocks for constructing clustered dot growth assemblies. Like the traditional Holladay dot description, the SRE halftone is characterized by a size, height, and shift, all of which are specified at the lower resolution. Each low resolution pixel position in the SRE halftone brick contains a pair of lists. The first of these is a list of digital thresholds at which a transition in SRE patterns occurs for that pixel position, and the second is the corresponding list of SRE codes. For normal cluster dot growth sequences, this provides a simple and compact mechanism for specifying higher resolution halftones. Techniques for emulating traditional high resolution Holladay dots using SREH are discussed, including mechanisms for choosing substitutions for patterns that do not exist among the available SRE patterns.

The hybrid screen is a halftoning method that generates stochastic dispersed dot textures in highlights and periodic clustered dot textures in midtones. Each tone level is sequentially designed from highlight to midtone by applying an iterative halftoning algorithm such as direct binary search (DBS). By allowing random seeding and swap-only DBS in a predefined core region within each microcell, we design each level while satisfying the stacking constraint and guaranteeing a smooth transition between different levels. This paper focuses on a number of enhancements to the original hybrid screen and their impacts on print quality. These include analytical determination of the human visual system filter in the spatial domain for DBS, multilevel screen design either by extending a bilevel screen or by directly generating a multilevel screen on the high resolution grid. Our results show that the multilevel screen design method has a direct impact on hybrid screen design parameters such as the optimal core size. We also extend the whole design process to color by jointly optimizing the color screens using color DBS. Our results demonstrate a significant improvement in the highlights over halftones generated by independently designed screen.

A spatially-adaptive method for color printing is proposed that is robust to printer instabilities, reproduces smooth regions with the quality of ordered dither, reproduces sharp edges significantly better than ordered dither, and may be less susceptible to moire. The new method acts in parallel on square, non-overlapping blocks of each color plane of the image. For blocks with low spatial activity, standard ordered dither is used, which ensures that smooth regions are printed with acceptable quality. Blocks with high spatial activity are halftoned with a proposed variant of dither, called ranked dither. Ranked dither uses the the same ordered dither matrix as standard dither, but the ranks of the thresholds are used rather than the thresholds themselves. Ranked dither is more sensitive than ordered dither to edges and more accurately reproduces sharp edges. Experiments were done with standard ordered dither masks of size 130, 130, 128, 144 for the cyan, magenta, yellow, and black planes respectively. Both on-screen and in-print, the results were sharper halftones. The entire process can be implemented in parallel and is not computationally expensive.

We present a specialized form of error diffusion that addresses certain long-standing problems associated with operating on images possessing halftone structure as well as other images with local high contrast. For instance, when rendering an image to printable form via quantization reduction, image quality defects often result if that image is a scanned halftone. Rendering such an image via conventional error diffusion typically produces fragmented dots, which can appear grainy and be unstable in printed density. Rendering by simple thresholding or rehalftoning often produces moire, and descreening blurs the image. Another difficulty arises in printers that utilize a binary image path, where an image is rasterized directly to halftone form. In that form it is very difficult to perform basic image processing operations such as applying a digital tone reproduction curve. The image processing operator introduced in this paper, rank-order error diffusion (ROED), has been developed to address these problems. ROED utilizes brightness ranking of pixels within a diffusion mask to diffuse quantization error at a pixel. This approach to diffusion results in an imagestructure- adaptive quantization with many useful properties. The present paper describes the basic methodology of ROED as well as several applications.

FM halftoning generates good tone rendition but it is not appropriate for electrophotographic (EP) printers due to the inherent instability of the EP process. Although AM halftoning yields stable dots, it is susceptible to moire and contouring artifacts. To combine the strengths of AM and FM halftoning, the AM/FM halftoning algorithm exploits each advantage of AM and FM halftoning. The resulting halftone textures have green noise spectral characteristics. In this paper, we present an improved training procedure for the AM/FM halftoning algorithm. Since most of the green noise energy is concentrated in the middle frequencies, the tone dependent error diffusion (TDED) parameters (weights and thresholds) are optimized using a new cost function with normalization to distribute the cost evenly over all frequencies. With the new cost function, we can obtain image quality that is very close to the direct binary search (DBS) search-based dispersed-dot halftoning algorithm. The cost function for training the AM part is also modified by penalizing variation in measured tone value across the multiple printer conditions for each combination of dot size and dot density.

Recently, speed and resolution of electrophotographic printer engine have been significantly improved. In today's market, it is not difficult to find low to mid-end electrophotographic printers with the spatial resolution greater than 600 dpi and/or bit-depth resolution more than 1 bit. Printing speed is determined by the processing time at computer, data transmission time between computer and printer, and processing and printing time at printer. When halftoning is performed at computer side, halftoned data would be compressed and sent to printer. In this case, increase in the spatial and bit-depth resolution would increase data size to be transmitted and memory size at printer. For a high-speed printer, increased transmission time may limit the throughput in imaging chain. One of possible solutions to this problem is to develop resolution enhancement techniques. In this paper, a fast and efficient spatial resolution enhancement technique is proposed. Objectives of the proposed technique are to reduce the data size for transmission and minimize image quality deterioration. In the proposed technique, number of black pixels in the halftoned data is binary coded for data reduction. At printer, black pixel placement algorithm is applied to binary coded data. For non-edge area, screen order is utilized for the black pixel placement. When identified as edge area, locations of black pixels are selected by the edge order designed by genetic algorithm.

Many new proposals are continually published in the halftoning domain. Alas, the demonstration of the interest of the proposed methods is often limited to a few favourable tests for the proposed methods, and images showing the defects of the other halftoning methods. The halftoning community needs to be able to compare a halftoning method with the innovations that appear in this domain. A complete and measured evaluation of quality is necessary through to a well defined set of test images and metrics to evaluate the algorithm. This paper proposes a protocol for the quality assessment of digital halftoning algorithm that can be used to compare one algorithm to another. It discusses the assessment of halftoner quality. It analyzes the perceived image quality concepts and defines the technical criteria that a good halftoner must match. A first sketch of a simple quality assessment protocol is proposed. It is composed of test images and quality metrics. This protocol could be used to provide new proposed halftoning algorithms with objective results.

By selecting halftone frequencies from high-order harmonics of two common rosette fundamentals for all color separations, a true moiré-free color halftoning can be achieved. With such screen configurations, the interference between any two frequency components, fundamentals or high-order harmonics, of different colors will also result in a linear combination of the two rosette fundamentals. Thereby, no visible interference, or moiré, at all will be shown in the output. The halftone outputs are two-dimensionally repeated patterns, as visually pleasant uniform rosettes. The uniform-rosette configurations can be implemented by single-cell non-orthogonal halftone screens for digital halftoning. Unlike "dot-on-dot" screening, or using one screen for all colors, uniform-rosette halftoning is robust to mis-registration between color separations. Several designs of uniform-rosette halftone screens have been successfully applied to Xerox iGen3 color printers for high-quality color reproduction.

In H.264 video coding standard, the combination encoding frame was adopted. It introduces some new algorithms, and modifies several aspects of the encoding scheme. So the encoding scheme improves the encoding efficiency obviously. But the H.264 standard is not supporting FGS encoding. So a H.264 based self-adaptive FGS (Fine Granular Scalable)(H.264-FGS) encoding scheme is proposed in this paper. In this encoding scheme, the base layer of encoder is keeping H.264 encoder architecture, which consists of the motion estimation, motion compensation, intra predictive, integer transformation, loop filtering, content based arithmetic encoding, and etc. In the base layer generated block we obtain base code flux of FGS. Subtracting the original image from the reconstruction image of the base layer, we get the residual error. Then after the DCT transform and the variable length encoding compresses, we obtain the enhanced code flux of FGS. Compared with the original MPEG-4 FGS encoding scheme, the proposed FGS encoding scheme has the feature of increasing encoding efficiency by 1~3 dB and keep the all properties that MPEG-4 FGS encoding technology provided.

The paper discussed the digital watermarking algorithm which embeds watermarking in halftone image in the process of halftoning. The digital watermarking algorithm based on human vision system (HVS) model which can minimize the visual error between the embedded watermarking halftone image and the original continuous-tone image using iterative binary search method. The algorithm can embed large amount of information in halftone image under the precondition of watermarking invisibility; on the other hand, the extraction of watermarking is reliable. All experiments indicate that this algorithm can embed more information than other algorithm and has strong robustness, so the algorithm can resist unconsciously or intentionally attacks which caused by printing, scan, dirty, clip, etc., at the same time, the watermarking embedded in halftone image using the algorithms has the advantage of invisibility, high vision quality of halftone image.

A strategy for to compress color images in digital holograms of gray tones was developed. The procedure codifies the information of each channel of the RGB model in a system of fringes, it is a gray image denominated "hologram". The fringes in their intensity of gray tone carry the signal of the channel, in this manner the amplitude information for each channel of color image is stored. The angles of fringes define how the information of each channel will be packaged. The sum of the different gray fringes images is the hologram, it is the "object" for a digital holographic system. The RGB channels are high intensity peaks of information in the hologram's Fourier space, and when the peaks are filtered each channel can be extracted. Parameters such as: space frequency, visibility, direction and quality of the fringes affect the quality of the reconstructed image. However, the propose methodology allow a radius 3:1 for the compression of color image, too with this process is possible the compression of different spectrum in a one color image.